diff --git a/data/workflows.json b/data/workflows.json
index bbc48ae81c1c2ddc97415a98015c2078a4047e0a..3440a0af2fe3a3f3907194da3cd7bbdf1005861a 100644
--- a/data/workflows.json
+++ b/data/workflows.json
@@ -1,228 +1,345 @@
[
{
- "eval_workflow_id": "wf-datareichsanzeiger_tables-eval",
- "label": "Workflow on data reichsanzeiger_tables",
+ "eval_workflow_id": "wf-datablumenbach_anatomie_1805-eval",
+ "label": "Workflow on data blumenbach_anatomie_1805",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_tables.ocrd.zip",
- "label": "GT workspace reichsanzeiger_tables"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/blumenbach_anatomie_1805.ocrd.zip",
+ "label": "GT workspace blumenbach_anatomie_1805"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_tables_ocr.zip",
- "label": "OCR workspace for reichsanzeiger_tables"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/blumenbach_anatomie_1805_ocr.zip",
+ "label": "OCR workspace for blumenbach_anatomie_1805"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_tables_evaluation.zip",
- "label": "Evaluation workspace for reichsanzeiger_tables"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/blumenbach_anatomie_1805_evaluation.zip",
+ "label": "Evaluation workspace for blumenbach_anatomie_1805"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
- "find_tables": true,
- "model": "Fraktur_GT4HistOCR",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
"dpi": 0,
- "padding": 0,
- "overwrite_segments": false,
- "overwrite_text": true,
- "shrink_polygons": false,
- "block_polygons": false,
- "find_staves": false,
- "sparse_text": false,
- "raw_lines": false,
- "char_whitelist": "",
- "char_blacklist": "",
- "char_unblacklist": "",
- "tesseract_parameters": {},
- "xpath_parameters": {},
- "xpath_model": {},
- "auto_model": false,
- "oem": "DEFAULT"
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
}
}
],
- "workflow_model": "Fraktur_GT4HistOCR",
- "eval_tool": "ocrd-dinglehopper vNone",
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua"
+ "Antiqua",
+ "Fraktur"
],
- "publication_century": "1820-1939",
+ "publication_century": "1800-1900",
"publication_decade": "",
"publication_year": "19th century",
- "number_of_pages": 5,
- "layout": "reichsanzeiger-gt"
+ "number_of_pages": 3,
+ "layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 530.564159,
- "cpu_time": 724.869668,
- "cer_mean": 1.479972651687655,
- "cer_median": 1.331401349741961,
+ "wall_time": 25.104598,
+ "cpu_time": 24.163375,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
"cer_range": [
- 1.090097148229395,
- 1.9778823058446757
+ 1.0,
+ 1.0
],
- "cer_standard_deviation": 0.3854879135899152,
- "wer": 1.7910299442503619,
- "pages_per_minute": 0.5654358571175178
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 7.170001288210232
},
"by_page": [
{
- "page_id": "P_1877_7_0059",
- "cer": 1.090097148229395,
- "wer": 1.4218399401645474
- },
- {
- "page_id": "P_1883_55_0044",
- "cer": 1.2073319135990264,
- "wer": 1.4705882352941178
- },
- {
- "page_id": "P_1929_250_0019",
- "cer": 1.331401349741961,
- "wer": 1.7407221664994985
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
},
{
- "page_id": "P_1932_300_0488",
- "cer": 1.9778823058446757,
- "wer": 2.349493487698987
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
},
{
- "page_id": "P_1936_123_0292",
- "cer": 1.7931505410232167,
- "wer": 1.9725058915946583
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
}
]
}
},
{
- "eval_workflow_id": "wf-datasilesius_seelenlust01_1657-eval",
- "label": "Workflow on data silesius_seelenlust01_1657",
+ "eval_workflow_id": "wf-databallenstedt_delatio_1777-eval",
+ "label": "Workflow on data ballenstedt_delatio_1777",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/silesius_seelenlust01_1657.ocrd.zip",
- "label": "GT workspace silesius_seelenlust01_1657"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/ballenstedt_delatio_1777.ocrd.zip",
+ "label": "GT workspace ballenstedt_delatio_1777"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/silesius_seelenlust01_1657_ocr.zip",
- "label": "OCR workspace for silesius_seelenlust01_1657"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/ballenstedt_delatio_1777_ocr.zip",
+ "label": "OCR workspace for ballenstedt_delatio_1777"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/silesius_seelenlust01_1657_evaluation.zip",
- "label": "Evaluation workspace for silesius_seelenlust01_1657"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/ballenstedt_delatio_1777_evaluation.zip",
+ "label": "Evaluation workspace for ballenstedt_delatio_1777"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
- "find_tables": true,
- "model": "Fraktur_GT4HistOCR",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
"dpi": 0,
- "padding": 0,
- "overwrite_segments": false,
- "overwrite_text": true,
- "shrink_polygons": false,
- "block_polygons": false,
- "find_staves": false,
- "sparse_text": false,
- "raw_lines": false,
- "char_whitelist": "",
- "char_blacklist": "",
- "char_unblacklist": "",
- "tesseract_parameters": {},
- "xpath_parameters": {},
- "xpath_model": {},
- "auto_model": false,
- "oem": "DEFAULT"
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
}
}
],
- "workflow_model": "Fraktur_GT4HistOCR",
- "eval_tool": "ocrd-dinglehopper vNone",
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
"document_metadata": {
"data_properties": {
"fonts": [
"Antiqua",
"Fraktur"
],
- "publication_century": "1600-1700",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "17th century",
- "number_of_pages": 5,
- "layout": "complex"
+ "publication_year": "18th century",
+ "number_of_pages": 3,
+ "layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 13.583684,
- "cpu_time": 17.039027,
- "cer_mean": 0.3007055286105995,
- "cer_median": 0.2951219512195122,
+ "wall_time": 31.663677000000007,
+ "cpu_time": 30.643176,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
"cer_range": [
- 0.19271623672230653,
- 0.44970414201183434
+ 1.0,
+ 1.0
],
- "cer_standard_deviation": 0.10657123719012947,
- "wer": 0.5174305966287508,
- "pages_per_minute": 22.08531941703002
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 5.6847472262933945
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.44970414201183434,
- "wer": 0.7101449275362319
- },
- {
- "page_id": "phys_0002",
- "cer": 0.20913884007029876,
- "wer": 0.5104166666666666
- },
- {
- "page_id": "phys_0003",
- "cer": 0.2951219512195122,
- "wer": 0.4647887323943662
+ "page_id": "phys_00003",
+ "cer": 1.0,
+ "wer": 1.0
},
{
- "page_id": "phys_0004",
- "cer": 0.19271623672230653,
- "wer": 0.3870967741935484
+ "page_id": "phys_00005",
+ "cer": 1.0,
+ "wer": 1.0
},
{
- "page_id": "phys_0005",
- "cer": 0.35684647302904565,
- "wer": 0.5147058823529411
+ "page_id": "phys_00010",
+ "cer": 1.0,
+ "wer": 1.0
}
]
}
},
{
- "eval_workflow_id": "wf-databenner_herrnhuterey04_1748-eval",
- "label": "Workflow on data benner_herrnhuterey04_1748",
+ "eval_workflow_id": "wf-datahuebner_handbuch_1696-eval",
+ "label": "Workflow on data huebner_handbuch_1696",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -233,16 +350,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/benner_herrnhuterey04_1748.ocrd.zip",
- "label": "GT workspace benner_herrnhuterey04_1748"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/huebner_handbuch_1696.ocrd.zip",
+ "label": "GT workspace huebner_handbuch_1696"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/benner_herrnhuterey04_1748_ocr.zip",
- "label": "OCR workspace for benner_herrnhuterey04_1748"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/huebner_handbuch_1696_ocr.zip",
+ "label": "OCR workspace for huebner_handbuch_1696"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/benner_herrnhuterey04_1748_evaluation.zip",
- "label": "Evaluation workspace for benner_herrnhuterey04_1748"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/huebner_handbuch_1696_evaluation.zip",
+ "label": "Evaluation workspace for huebner_handbuch_1696"
},
"workflow_steps": [
{
@@ -280,90 +397,154 @@
"Antiqua",
"Fraktur"
],
- "publication_century": "1700-1800",
+ "publication_century": "1600-1700",
"publication_decade": "",
- "publication_year": "18th century",
- "number_of_pages": 4,
+ "publication_year": "17th century",
+ "number_of_pages": 3,
"layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 19.971749,
- "cpu_time": 23.245253,
- "cer_mean": 0.20433470587543373,
- "cer_median": 0.1700173533179325,
+ "wall_time": 14.267812,
+ "cpu_time": 16.333578,
+ "cer_mean": 0.20627969569747484,
+ "cer_median": 0.08702290076335878,
"cer_range": [
- 0.04063745019920319,
- 0.43666666666666665
+ 0.08487084870848709,
+ 0.44694533762057875
],
- "cer_standard_deviation": 0.19097881579976753,
- "wer": 0.31449770246107817,
- "pages_per_minute": 12.016974577439363
+ "cer_standard_deviation": 0.20842533731220508,
+ "wer": 0.3271033769383102,
+ "pages_per_minute": 12.615809627993416
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.43666666666666665,
- "wer": 0.583941605839416
+ "cer": 0.44694533762057875,
+ "wer": 0.6341463414634146
},
{
"page_id": "phys_0002",
- "cer": 0.04063745019920319,
- "wer": 0.10344827586206896
+ "cer": 0.08702290076335878,
+ "wer": 0.1504424778761062
},
{
"page_id": "phys_0003",
- "cer": 0.05536912751677853,
- "wer": 0.18435754189944134
- },
- {
- "page_id": "phys_0004",
- "cer": 0.28466557911908646,
- "wer": 0.3862433862433862
+ "cer": 0.08487084870848709,
+ "wer": 0.19672131147540983
}
]
}
},
{
- "eval_workflow_id": "wf-databohse_helicon_1696-eval",
- "label": "Workflow on data bohse_helicon_1696",
+ "eval_workflow_id": "wf-datareichsanzeiger_title_pages-eval",
+ "label": "Workflow on data reichsanzeiger_title_pages",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/bohse_helicon_1696.ocrd.zip",
- "label": "GT workspace bohse_helicon_1696"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_title_pages.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_title_pages"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bohse_helicon_1696_ocr.zip",
- "label": "OCR workspace for bohse_helicon_1696"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_title_pages_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_title_pages"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bohse_helicon_1696_evaluation.zip",
- "label": "Evaluation workspace for bohse_helicon_1696"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_title_pages_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_title_pages"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-crop",
+ "params": {
+ "dpi": 0,
+ "padding": 4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
"find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "overwrite_segments": false,
+ "segmentation_level": "word",
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -383,194 +564,335 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua",
- "Fraktur"
+ "Antiqua"
],
- "publication_century": "1600-1700",
+ "publication_century": "1820-1939",
"publication_decade": "",
- "publication_year": "17th century",
+ "publication_year": "19th century",
"number_of_pages": 5,
- "layout": "simple"
+ "layout": "reichsanzeiger-gt"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 16.040643,
- "cpu_time": 19.265802,
- "cer_mean": 0.40403175304113403,
- "cer_median": 0.45517241379310347,
+ "wall_time": 2039.5017899999998,
+ "cpu_time": 3039.7441000000003,
+ "cer_mean": 0.503371617521129,
+ "cer_median": 0.413282477002468,
"cer_range": [
- 0.1955040871934605,
- 0.5079365079365079
+ 0.08001625727131861,
+ 1.0
],
- "cer_standard_deviation": 0.12207840266111031,
- "wer": 0.5356334009868995,
- "pages_per_minute": 18.70249216318822
+ "cer_standard_deviation": 0.3496032851993558,
+ "wer": 0.6522771332007985,
+ "pages_per_minute": 0.14709474709507367
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.5079365079365079,
- "wer": 0.6382978723404256
+ "page_id": "P_1881_115_0163",
+ "cer": 0.413282477002468,
+ "wer": 0.5798611111111112
},
{
- "page_id": "phys_0002",
- "cer": 0.40540540540540543,
- "wer": 0.5740740740740741
+ "page_id": "P_1885_5_0054",
+ "cer": 0.08001625727131861,
+ "wer": 0.22626162018592297
},
{
- "page_id": "phys_0003",
- "cer": 0.45517241379310347,
- "wer": 0.5467625899280576
+ "page_id": "P_1887_134_0444",
+ "cer": 1.0,
+ "wer": 1.0
},
{
- "page_id": "phys_0004",
- "cer": 0.45614035087719296,
- "wer": 0.6018518518518519
+ "page_id": "P_1916_169_0087",
+ "cer": 0.6772530470255524,
+ "wer": 0.8682910206168002
},
{
- "page_id": "phys_0005",
- "cer": 0.1955040871934605,
- "wer": 0.31718061674008813
+ "page_id": "P_1918_267_0129",
+ "cer": 0.3463063063063063,
+ "wer": 0.5869719140901581
}
]
}
},
{
- "eval_workflow_id": "wf-databuerger_gedichte_1778-eval",
- "label": "Workflow on data buerger_gedichte_1778",
+ "eval_workflow_id": "wf-dataarnimb_goethe03_1835-eval",
+ "label": "Workflow on data arnimb_goethe03_1835",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/buerger_gedichte_1778.ocrd.zip",
- "label": "GT workspace buerger_gedichte_1778"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/arnimb_goethe03_1835.ocrd.zip",
+ "label": "GT workspace arnimb_goethe03_1835"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/buerger_gedichte_1778_ocr.zip",
- "label": "OCR workspace for buerger_gedichte_1778"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/arnimb_goethe03_1835_ocr.zip",
+ "label": "OCR workspace for arnimb_goethe03_1835"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/buerger_gedichte_1778_evaluation.zip",
- "label": "Evaluation workspace for buerger_gedichte_1778"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/arnimb_goethe03_1835_evaluation.zip",
+ "label": "Evaluation workspace for arnimb_goethe03_1835"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
- "find_tables": true,
- "model": "Fraktur_GT4HistOCR",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
"dpi": 0,
- "padding": 0,
- "overwrite_segments": false,
- "overwrite_text": true,
- "shrink_polygons": false,
- "block_polygons": false,
- "find_staves": false,
- "sparse_text": false,
- "raw_lines": false,
- "char_whitelist": "",
- "char_blacklist": "",
- "char_unblacklist": "",
- "tesseract_parameters": {},
- "xpath_parameters": {},
- "xpath_model": {},
- "auto_model": false,
- "oem": "DEFAULT"
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
}
}
],
- "workflow_model": "Fraktur_GT4HistOCR",
- "eval_tool": "ocrd-dinglehopper vNone",
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua",
"Fraktur"
],
- "publication_century": "1700-1800",
+ "publication_century": "1800-1900",
"publication_decade": "",
- "publication_year": "18th century",
- "number_of_pages": 2,
- "layout": "complex"
+ "publication_year": "19th century",
+ "number_of_pages": 1,
+ "layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 5.921714,
- "cpu_time": 7.016172,
- "cer_mean": 0.10672693293515115,
- "cer_median": 0.10672693293515115,
+ "wall_time": 6.831029999999999,
+ "cpu_time": 6.5121720000000005,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
"cer_range": [
- 0.04484304932735426,
- 0.16861081654294804
+ 1.0,
+ 1.0
],
- "cer_standard_deviation": 0.08751702749046443,
- "wer": 0.23135901859306116,
- "pages_per_minute": 20.26440317786371
+ "cer_standard_deviation": null,
+ "wer": 1.0,
+ "pages_per_minute": 8.783448469703691
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.16861081654294804,
- "wer": 0.3546099290780142
- },
- {
- "page_id": "phys_0002",
- "cer": 0.04484304932735426,
- "wer": 0.10810810810810811
+ "cer": 1.0,
+ "wer": 1.0
}
]
}
},
{
- "eval_workflow_id": "wf-datarollenhagen_reysen_1603-eval",
- "label": "Workflow on data rollenhagen_reysen_1603",
+ "eval_workflow_id": "wf-datann_besuch_1780-eval",
+ "label": "Workflow on data nn_besuch_1780",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/rollenhagen_reysen_1603.ocrd.zip",
- "label": "GT workspace rollenhagen_reysen_1603"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/nn_besuch_1780.ocrd.zip",
+ "label": "GT workspace nn_besuch_1780"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/rollenhagen_reysen_1603_ocr.zip",
- "label": "OCR workspace for rollenhagen_reysen_1603"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/nn_besuch_1780_ocr.zip",
+ "label": "OCR workspace for nn_besuch_1780"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/rollenhagen_reysen_1603_evaluation.zip",
- "label": "Evaluation workspace for rollenhagen_reysen_1603"
- },
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/nn_besuch_1780_evaluation.zip",
+ "label": "Evaluation workspace for nn_besuch_1780"
+ },
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
"find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "overwrite_segments": false,
+ "segmentation_level": "word",
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -593,85 +915,170 @@
"Antiqua",
"Fraktur"
],
- "publication_century": "1600-1700",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "17th century",
- "number_of_pages": 3,
- "layout": "simple"
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 16.777821,
- "cpu_time": 19.16547,
- "cer_mean": 0.21046220070684576,
- "cer_median": 0.16929133858267717,
+ "wall_time": 42.463463,
+ "cpu_time": 61.604054,
+ "cer_mean": 0.07900170040044713,
+ "cer_median": 0.018462174141687326,
"cer_range": [
- 0.14512471655328799,
- 0.3169705469845722
+ 0.013259668508287293,
+ 0.26582278481012656
],
- "cer_standard_deviation": 0.093027024434784,
- "wer": 0.3190752126565147,
- "pages_per_minute": 10.728449183001775
+ "cer_standard_deviation": 0.1245848990441196,
+ "wer": 0.13667696498571524,
+ "pages_per_minute": 5.651917743967326
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.3169705469845722,
- "wer": 0.4649122807017544
+ "page_id": "phys_00001",
+ "cer": 0.26582278481012656,
+ "wer": 0.3620689655172414
},
{
- "page_id": "phys_0002",
- "cer": 0.16929133858267717,
- "wer": 0.28104575163398693
+ "page_id": "phys_00002",
+ "cer": 0.013259668508287293,
+ "wer": 0.057803468208092484
},
{
- "page_id": "phys_0003",
- "cer": 0.14512471655328799,
- "wer": 0.2112676056338028
+ "page_id": "phys_00003",
+ "cer": 0.020697167755991286,
+ "wer": 0.06936416184971098
+ },
+ {
+ "page_id": "phys_00004",
+ "cer": 0.016227180527383367,
+ "wer": 0.05747126436781609
}
]
}
},
{
- "eval_workflow_id": "wf-dataeuler_rechenkunst01_1738-eval",
- "label": "Workflow on data euler_rechenkunst01_1738",
+ "eval_workflow_id": "wf-databernd_lebensbeschreibung_1738-eval",
+ "label": "Workflow on data bernd_lebensbeschreibung_1738",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/euler_rechenkunst01_1738.ocrd.zip",
- "label": "GT workspace euler_rechenkunst01_1738"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/bernd_lebensbeschreibung_1738.ocrd.zip",
+ "label": "GT workspace bernd_lebensbeschreibung_1738"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/euler_rechenkunst01_1738_ocr.zip",
- "label": "OCR workspace for euler_rechenkunst01_1738"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bernd_lebensbeschreibung_1738_ocr.zip",
+ "label": "OCR workspace for bernd_lebensbeschreibung_1738"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/euler_rechenkunst01_1738_evaluation.zip",
- "label": "Evaluation workspace for euler_rechenkunst01_1738"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bernd_lebensbeschreibung_1738_evaluation.zip",
+ "label": "Evaluation workspace for bernd_lebensbeschreibung_1738"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
"find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "overwrite_segments": false,
+ "segmentation_level": "word",
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -697,97 +1104,162 @@
"publication_century": "1700-1800",
"publication_decade": "",
"publication_year": "18th century",
- "number_of_pages": 6,
+ "number_of_pages": 3,
"layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 18.514995,
- "cpu_time": 23.577633,
- "cer_mean": 0.25835849983393794,
- "cer_median": 0.230420483908856,
+ "wall_time": 30.325404000000006,
+ "cpu_time": 46.848454,
+ "cer_mean": 0.08856691777926955,
+ "cer_median": 0.03746177370030581,
"cer_range": [
- 0.08586296617519515,
- 0.5747368421052632
+ 0.0290519877675841,
+ 0.1991869918699187
],
- "cer_standard_deviation": 0.16915619139008603,
- "wer": 0.37827288927088376,
- "pages_per_minute": 19.443699552713895
+ "cer_standard_deviation": 0.09589203157053293,
+ "wer": 0.14344277534022457,
+ "pages_per_minute": 5.935617543627777
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.08586296617519515,
- "wer": 0.10465116279069768
+ "cer": 0.1991869918699187,
+ "wer": 0.27941176470588236
},
{
"page_id": "phys_0002",
- "cer": 0.5747368421052632,
- "wer": 0.7987804878048781
+ "cer": 0.0290519877675841,
+ "wer": 0.05357142857142857
},
{
"page_id": "phys_0003",
- "cer": 0.2767102229054573,
- "wer": 0.3786407766990291
- },
- {
- "page_id": "phys_0004",
- "cer": 0.22828282828282828,
- "wer": 0.27741935483870966
- },
- {
- "page_id": "phys_0005",
- "cer": 0.152,
- "wer": 0.27218934911242604
- },
- {
- "page_id": "phys_0006",
- "cer": 0.23255813953488372,
- "wer": 0.43795620437956206
+ "cer": 0.03746177370030581,
+ "wer": 0.09734513274336283
}
]
}
},
{
- "eval_workflow_id": "wf-datakistler_kraeuter_1500-eval",
- "label": "Workflow on data kistler_kraeuter_1500",
+ "eval_workflow_id": "wf-dataluther_auszlegunge_1520-eval",
+ "label": "Workflow on data luther_auszlegunge_1520",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/kistler_kraeuter_1500.ocrd.zip",
- "label": "GT workspace kistler_kraeuter_1500"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/luther_auszlegunge_1520.ocrd.zip",
+ "label": "GT workspace luther_auszlegunge_1520"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/kistler_kraeuter_1500_ocr.zip",
- "label": "OCR workspace for kistler_kraeuter_1500"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luther_auszlegunge_1520_ocr.zip",
+ "label": "OCR workspace for luther_auszlegunge_1520"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/kistler_kraeuter_1500_evaluation.zip",
- "label": "Evaluation workspace for kistler_kraeuter_1500"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luther_auszlegunge_1520_evaluation.zip",
+ "label": "Evaluation workspace for luther_auszlegunge_1520"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
"find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "overwrite_segments": false,
+ "segmentation_level": "word",
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -820,35 +1292,35 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 7.833456,
- "cpu_time": 9.538595,
- "cer_mean": 0.1809783660358373,
- "cer_median": 0.1809783660358373,
+ "wall_time": 27.237084000000003,
+ "cpu_time": 36.259029,
+ "cer_mean": 0.5161379724470825,
+ "cer_median": 0.5161379724470825,
"cer_range": [
- 0.13479623824451412,
- 0.2271604938271605
+ 0.0754414125200642,
+ 0.9568345323741008
],
- "cer_standard_deviation": 0.06531139146173669,
- "wer": 0.408373786407767,
- "pages_per_minute": 15.318909048573197
+ "cer_standard_deviation": 0.6232390519399567,
+ "wer": 0.6192780337941628,
+ "pages_per_minute": 4.405757973210347
},
"by_page": [
{
- "page_id": "phys_0007",
- "cer": 0.2271604938271605,
- "wer": 0.441747572815534
+ "page_id": "phys_0003",
+ "cer": 0.9568345323741008,
+ "wer": 0.9666666666666667
},
{
- "page_id": "phys_0021",
- "cer": 0.13479623824451412,
- "wer": 0.375
+ "page_id": "phys_0029",
+ "cer": 0.0754414125200642,
+ "wer": 0.271889400921659
}
]
}
},
{
- "eval_workflow_id": "wf-datacalvi_beutelschneider01_1627-eval",
- "label": "Workflow on data calvi_beutelschneider01_1627",
+ "eval_workflow_id": "wf-databohse_helicon_1696-eval",
+ "label": "Workflow on data bohse_helicon_1696",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -859,16 +1331,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/calvi_beutelschneider01_1627.ocrd.zip",
- "label": "GT workspace calvi_beutelschneider01_1627"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/bohse_helicon_1696.ocrd.zip",
+ "label": "GT workspace bohse_helicon_1696"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/calvi_beutelschneider01_1627_ocr.zip",
- "label": "OCR workspace for calvi_beutelschneider01_1627"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bohse_helicon_1696_ocr.zip",
+ "label": "OCR workspace for bohse_helicon_1696"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/calvi_beutelschneider01_1627_evaluation.zip",
- "label": "Evaluation workspace for calvi_beutelschneider01_1627"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bohse_helicon_1696_evaluation.zip",
+ "label": "Evaluation workspace for bohse_helicon_1696"
},
"workflow_steps": [
{
@@ -909,82 +1381,172 @@
"publication_century": "1600-1700",
"publication_decade": "",
"publication_year": "17th century",
- "number_of_pages": 3,
+ "number_of_pages": 5,
"layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 7.37275,
- "cpu_time": 9.77562,
- "cer_mean": 0.08254540344028655,
- "cer_median": 0.09331797235023041,
+ "wall_time": 15.131289,
+ "cpu_time": 18.178699,
+ "cer_mean": 0.40403175304113403,
+ "cer_median": 0.45517241379310347,
"cer_range": [
- 0.05025996533795494,
- 0.1040582726326743
+ 0.1955040871934605,
+ 0.5079365079365079
],
- "cer_standard_deviation": 0.02847104928930229,
- "wer": 0.1821206964037463,
- "pages_per_minute": 24.41422806958055
+ "cer_standard_deviation": 0.12207840266111031,
+ "wer": 0.5356334009868995,
+ "pages_per_minute": 19.826466866107705
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.1040582726326743,
- "wer": 0.21656050955414013
+ "cer": 0.5079365079365079,
+ "wer": 0.6382978723404256
},
{
"page_id": "phys_0002",
- "cer": 0.09331797235023041,
- "wer": 0.20689655172413793
+ "cer": 0.40540540540540543,
+ "wer": 0.5740740740740741
},
{
"page_id": "phys_0003",
- "cer": 0.05025996533795494,
- "wer": 0.12290502793296089
+ "cer": 0.45517241379310347,
+ "wer": 0.5467625899280576
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.45614035087719296,
+ "wer": 0.6018518518518519
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 0.1955040871934605,
+ "wer": 0.31718061674008813
}
]
}
},
{
- "eval_workflow_id": "wf-dataarnimb_goethe03_1835-eval",
- "label": "Workflow on data arnimb_goethe03_1835",
+ "eval_workflow_id": "wf-datacalvi_beutelschneider01_1627-eval",
+ "label": "Workflow on data calvi_beutelschneider01_1627",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/arnimb_goethe03_1835.ocrd.zip",
- "label": "GT workspace arnimb_goethe03_1835"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/calvi_beutelschneider01_1627.ocrd.zip",
+ "label": "GT workspace calvi_beutelschneider01_1627"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/arnimb_goethe03_1835_ocr.zip",
- "label": "OCR workspace for arnimb_goethe03_1835"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/calvi_beutelschneider01_1627_ocr.zip",
+ "label": "OCR workspace for calvi_beutelschneider01_1627"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/arnimb_goethe03_1835_evaluation.zip",
- "label": "Evaluation workspace for arnimb_goethe03_1835"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/calvi_beutelschneider01_1627_evaluation.zip",
+ "label": "Evaluation workspace for calvi_beutelschneider01_1627"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
"find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "overwrite_segments": false,
+ "segmentation_level": "word",
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -1004,143 +1566,217 @@
"document_metadata": {
"data_properties": {
"fonts": [
+ "Antiqua",
"Fraktur"
],
- "publication_century": "1800-1900",
+ "publication_century": "1600-1700",
"publication_decade": "",
- "publication_year": "19th century",
- "number_of_pages": 1,
+ "publication_year": "17th century",
+ "number_of_pages": 3,
"layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 4.707975,
- "cpu_time": 5.426059,
- "cer_mean": 0.004721435316336166,
- "cer_median": 0.004721435316336166,
+ "wall_time": 37.645064000000005,
+ "cpu_time": 55.114228999999995,
+ "cer_mean": 0.15501219595245352,
+ "cer_median": 0.11654526534859522,
"cer_range": [
- 0.004721435316336166,
- 0.004721435316336166
+ 0.09618717504332755,
+ 0.2523041474654378
],
- "cer_standard_deviation": null,
- "wer": 0.015873015873015872,
- "pages_per_minute": 12.744332754528221
+ "cer_standard_deviation": 0.08486993479509113,
+ "wer": 0.360948415946103,
+ "pages_per_minute": 4.781503359909283
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.004721435316336166,
- "wer": 0.015873015873015872
+ "cer": 0.11654526534859522,
+ "wer": 0.46496815286624205
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.2523041474654378,
+ "wer": 0.4
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.09618717504332755,
+ "wer": 0.21787709497206703
}
]
}
},
{
- "eval_workflow_id": "wf-datablumenbach_anatomie_1805-eval",
- "label": "Workflow on data blumenbach_anatomie_1805",
+ "eval_workflow_id": "wf-datakistler_kraeuter_1500-eval",
+ "label": "Workflow on data kistler_kraeuter_1500",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/blumenbach_anatomie_1805.ocrd.zip",
- "label": "GT workspace blumenbach_anatomie_1805"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/kistler_kraeuter_1500.ocrd.zip",
+ "label": "GT workspace kistler_kraeuter_1500"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/blumenbach_anatomie_1805_ocr.zip",
- "label": "OCR workspace for blumenbach_anatomie_1805"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/kistler_kraeuter_1500_ocr.zip",
+ "label": "OCR workspace for kistler_kraeuter_1500"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/blumenbach_anatomie_1805_evaluation.zip",
- "label": "Evaluation workspace for blumenbach_anatomie_1805"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/kistler_kraeuter_1500_evaluation.zip",
+ "label": "Evaluation workspace for kistler_kraeuter_1500"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
- "find_tables": true,
- "model": "Fraktur_GT4HistOCR",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
"dpi": 0,
- "padding": 0,
- "overwrite_segments": false,
- "overwrite_text": true,
- "shrink_polygons": false,
- "block_polygons": false,
- "find_staves": false,
- "sparse_text": false,
- "raw_lines": false,
- "char_whitelist": "",
- "char_blacklist": "",
- "char_unblacklist": "",
- "tesseract_parameters": {},
- "xpath_parameters": {},
- "xpath_model": {},
- "auto_model": false,
- "oem": "DEFAULT"
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
}
}
],
- "workflow_model": "Fraktur_GT4HistOCR",
- "eval_tool": "ocrd-dinglehopper vNone",
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
"document_metadata": {
"data_properties": {
"fonts": [
"Antiqua",
"Fraktur"
],
- "publication_century": "1800-1900",
+ "publication_century": "1500-1600",
"publication_decade": "",
- "publication_year": "19th century",
- "number_of_pages": 3,
+ "publication_year": "16th century",
+ "number_of_pages": 2,
"layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 13.404813,
- "cpu_time": 15.520204,
- "cer_mean": 0.08328200324172261,
- "cer_median": 0.08736842105263158,
+ "wall_time": 14.867348999999999,
+ "cpu_time": 14.321257,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
"cer_range": [
- 0.04055496264674493,
- 0.12192262602579132
+ 1.0,
+ 1.0
],
- "cer_standard_deviation": 0.04083746158658049,
- "wer": 0.23519468186134854,
- "pages_per_minute": 13.428012759297722
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 8.071378427990089
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.08736842105263158,
- "wer": 0.22666666666666666
- },
- {
- "page_id": "phys_0002",
- "cer": 0.04055496264674493,
- "wer": 0.14814814814814814
+ "page_id": "phys_0007",
+ "cer": 1.0,
+ "wer": 1.0
},
{
- "page_id": "phys_0003",
- "cer": 0.12192262602579132,
- "wer": 0.33076923076923076
+ "page_id": "phys_0021",
+ "cer": 1.0,
+ "wer": 1.0
}
]
}
},
{
- "eval_workflow_id": "wf-datareichsanzeiger_title_pages-eval",
- "label": "Workflow on data reichsanzeiger_title_pages",
+ "eval_workflow_id": "wf-dataeuler_rechenkunst01_1738-eval",
+ "label": "Workflow on data euler_rechenkunst01_1738",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -1151,16 +1787,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_title_pages.ocrd.zip",
- "label": "GT workspace reichsanzeiger_title_pages"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/euler_rechenkunst01_1738.ocrd.zip",
+ "label": "GT workspace euler_rechenkunst01_1738"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_title_pages_ocr.zip",
- "label": "OCR workspace for reichsanzeiger_title_pages"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/euler_rechenkunst01_1738_ocr.zip",
+ "label": "OCR workspace for euler_rechenkunst01_1738"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_title_pages_evaluation.zip",
- "label": "Evaluation workspace for reichsanzeiger_title_pages"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/euler_rechenkunst01_1738_evaluation.zip",
+ "label": "Evaluation workspace for euler_rechenkunst01_1738"
},
"workflow_steps": [
{
@@ -1195,97 +1831,183 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua"
+ "Antiqua",
+ "Fraktur"
],
- "publication_century": "1820-1939",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "19th century",
- "number_of_pages": 5,
- "layout": "reichsanzeiger-gt"
+ "publication_year": "18th century",
+ "number_of_pages": 6,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 470.537927,
- "cpu_time": 603.540066,
- "cer_mean": 0.5074594248191785,
- "cer_median": 0.3389393598117977,
+ "wall_time": 16.461667,
+ "cpu_time": 21.693217,
+ "cer_mean": 0.25835849983393794,
+ "cer_median": 0.230420483908856,
"cer_range": [
- 0.09154875911296263,
- 1.0
+ 0.08586296617519515,
+ 0.5747368421052632
],
- "cer_standard_deviation": 0.36783069273412317,
- "wer": 0.630009265947987,
- "pages_per_minute": 0.6375681593037663
+ "cer_standard_deviation": 0.16915619139008603,
+ "wer": 0.37827288927088376,
+ "pages_per_minute": 21.86898811645261
},
"by_page": [
{
- "page_id": "P_1881_115_0163",
- "cer": 0.3377672361293631,
- "wer": 0.4791666666666667
+ "page_id": "phys_0001",
+ "cer": 0.08586296617519515,
+ "wer": 0.10465116279069768
},
{
- "page_id": "P_1885_5_0054",
- "cer": 0.09154875911296263,
- "wer": 0.20966135458167331
+ "page_id": "phys_0002",
+ "cer": 0.5747368421052632,
+ "wer": 0.7987804878048781
},
{
- "page_id": "P_1887_134_0444",
- "cer": 1.0,
- "wer": 1.0
+ "page_id": "phys_0003",
+ "cer": 0.2767102229054573,
+ "wer": 0.3786407766990291
},
{
- "page_id": "P_1916_169_0087",
- "cer": 0.3389393598117977,
- "wer": 0.5072414380644062
+ "page_id": "phys_0004",
+ "cer": 0.22828282828282828,
+ "wer": 0.27741935483870966
},
{
- "page_id": "P_1918_267_0129",
- "cer": 0.769041769041769,
- "wer": 0.953976870427189
+ "page_id": "phys_0005",
+ "cer": 0.152,
+ "wer": 0.27218934911242604
+ },
+ {
+ "page_id": "phys_0006",
+ "cer": 0.23255813953488372,
+ "wer": 0.43795620437956206
}
]
}
},
{
- "eval_workflow_id": "wf-dataluther_auszlegunge_1520-eval",
- "label": "Workflow on data luther_auszlegunge_1520",
+ "eval_workflow_id": "wf-databenner_herrnhuterey04_1748-eval",
+ "label": "Workflow on data benner_herrnhuterey04_1748",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/luther_auszlegunge_1520.ocrd.zip",
- "label": "GT workspace luther_auszlegunge_1520"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/benner_herrnhuterey04_1748.ocrd.zip",
+ "label": "GT workspace benner_herrnhuterey04_1748"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luther_auszlegunge_1520_ocr.zip",
- "label": "OCR workspace for luther_auszlegunge_1520"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/benner_herrnhuterey04_1748_ocr.zip",
+ "label": "OCR workspace for benner_herrnhuterey04_1748"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luther_auszlegunge_1520_evaluation.zip",
- "label": "Evaluation workspace for luther_auszlegunge_1520"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/benner_herrnhuterey04_1748_evaluation.zip",
+ "label": "Evaluation workspace for benner_herrnhuterey04_1748"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
"find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "overwrite_segments": false,
+ "segmentation_level": "word",
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -1308,45 +2030,55 @@
"Antiqua",
"Fraktur"
],
- "publication_century": "1500-1600",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "16th century",
- "number_of_pages": 2,
- "layout": "simple"
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 10.160861,
- "cpu_time": 11.973182,
- "cer_mean": 0.24855845660550213,
- "cer_median": 0.24855845660550213,
+ "wall_time": 58.189493,
+ "cpu_time": 87.823723,
+ "cer_mean": 0.16726583056278752,
+ "cer_median": 0.09637318392327315,
"cer_range": [
- 0.07865168539325842,
- 0.4184652278177458
+ 0.03187250996015936,
+ 0.4444444444444444
],
- "cer_standard_deviation": 0.2402844601873776,
- "wer": 0.37300307219662054,
- "pages_per_minute": 11.81002279235982
+ "cer_standard_deviation": 0.18889822286887584,
+ "wer": 0.28912998545359864,
+ "pages_per_minute": 4.12445593915039
},
"by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.4444444444444444,
+ "wer": 0.6204379562043796
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.03187250996015936,
+ "wer": 0.10919540229885058
+ },
{
"page_id": "phys_0003",
- "cer": 0.4184652278177458,
- "wer": 0.48333333333333334
+ "cer": 0.1266778523489933,
+ "wer": 0.2681564245810056
},
{
- "page_id": "phys_0029",
- "cer": 0.07865168539325842,
- "wer": 0.2626728110599078
+ "page_id": "phys_0004",
+ "cer": 0.06606851549755302,
+ "wer": 0.15873015873015872
}
]
}
},
{
- "eval_workflow_id": "wf-datahuebner_handbuch_1696-eval",
- "label": "Workflow on data huebner_handbuch_1696",
+ "eval_workflow_id": "wf-databernd_lebensbeschreibung_1738-eval",
+ "label": "Workflow on data bernd_lebensbeschreibung_1738",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -1357,16 +2089,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/huebner_handbuch_1696.ocrd.zip",
- "label": "GT workspace huebner_handbuch_1696"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/bernd_lebensbeschreibung_1738.ocrd.zip",
+ "label": "GT workspace bernd_lebensbeschreibung_1738"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/huebner_handbuch_1696_ocr.zip",
- "label": "OCR workspace for huebner_handbuch_1696"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bernd_lebensbeschreibung_1738_ocr.zip",
+ "label": "OCR workspace for bernd_lebensbeschreibung_1738"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/huebner_handbuch_1696_evaluation.zip",
- "label": "Evaluation workspace for huebner_handbuch_1696"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bernd_lebensbeschreibung_1738_evaluation.zip",
+ "label": "Evaluation workspace for bernd_lebensbeschreibung_1738"
},
"workflow_steps": [
{
@@ -1404,9 +2136,9 @@
"Antiqua",
"Fraktur"
],
- "publication_century": "1600-1700",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "17th century",
+ "publication_year": "18th century",
"number_of_pages": 3,
"layout": "complex"
}
@@ -1414,146 +2146,207 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 15.981865,
- "cpu_time": 18.106327,
- "cer_mean": 0.20627969569747484,
- "cer_median": 0.08702290076335878,
+ "wall_time": 7.753105,
+ "cpu_time": 9.89478,
+ "cer_mean": 0.1790109644215708,
+ "cer_median": 0.05504587155963303,
"cer_range": [
- 0.08487084870848709,
- 0.44694533762057875
+ 0.008409785932721712,
+ 0.4735772357723577
],
- "cer_standard_deviation": 0.20842533731220508,
- "wer": 0.3271033769383102,
- "pages_per_minute": 11.262765640931143
+ "cer_standard_deviation": 0.2561653709691831,
+ "wer": 0.2391410103864555,
+ "pages_per_minute": 23.216504871274154
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.44694533762057875,
- "wer": 0.6341463414634146
+ "cer": 0.4735772357723577,
+ "wer": 0.5882352941176471
},
{
"page_id": "phys_0002",
- "cer": 0.08702290076335878,
- "wer": 0.1504424778761062
+ "cer": 0.05504587155963303,
+ "wer": 0.09821428571428571
},
{
"page_id": "phys_0003",
- "cer": 0.08487084870848709,
- "wer": 0.19672131147540983
+ "cer": 0.008409785932721712,
+ "wer": 0.030973451327433628
}
]
}
},
{
- "eval_workflow_id": "wf-datann_besuch_1780-eval",
- "label": "Workflow on data nn_besuch_1780",
+ "eval_workflow_id": "wf-datareichsanzeiger_tables-eval",
+ "label": "Workflow on data reichsanzeiger_tables",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/nn_besuch_1780.ocrd.zip",
- "label": "GT workspace nn_besuch_1780"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_tables.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_tables"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/nn_besuch_1780_ocr.zip",
- "label": "OCR workspace for nn_besuch_1780"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_tables_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_tables"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/nn_besuch_1780_evaluation.zip",
- "label": "Evaluation workspace for nn_besuch_1780"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_tables_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_tables"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
- "find_tables": true,
- "model": "Fraktur_GT4HistOCR",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
"dpi": 0,
- "padding": 0,
- "overwrite_segments": false,
- "overwrite_text": true,
- "shrink_polygons": false,
- "block_polygons": false,
- "find_staves": false,
- "sparse_text": false,
- "raw_lines": false,
- "char_whitelist": "",
- "char_blacklist": "",
- "char_unblacklist": "",
- "tesseract_parameters": {},
- "xpath_parameters": {},
- "xpath_model": {},
- "auto_model": false,
- "oem": "DEFAULT"
+ "level-of-operation": "page"
}
- }
- ],
- "workflow_model": "Fraktur_GT4HistOCR",
- "eval_tool": "ocrd-dinglehopper vNone",
- "document_metadata": {
- "data_properties": {
- "fonts": [
- "Antiqua",
- "Fraktur"
- ],
- "publication_century": "1700-1800",
+ },
+ {
+ "id": "ocrd-tesserocr-crop",
+ "params": {
+ "dpi": 0,
+ "padding": 4
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua"
+ ],
+ "publication_century": "1820-1939",
"publication_decade": "",
- "publication_year": "18th century",
- "number_of_pages": 4,
- "layout": "complex"
+ "publication_year": "19th century",
+ "number_of_pages": 5,
+ "layout": "reichsanzeiger-gt"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 14.316583,
- "cpu_time": 16.701302,
- "cer_mean": 0.06315407734401027,
- "cer_median": 0.026726016076928857,
+ "wall_time": 1761.044629,
+ "cpu_time": 1719.1173,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
"cer_range": [
- 0.01878453038674033,
- 0.18037974683544303
+ 1.0,
+ 1.0
],
- "cer_standard_deviation": 0.07825196427362012,
- "wer": 0.11662846322503488,
- "pages_per_minute": 16.76377666374721
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 0.17035343401282987
},
"by_page": [
{
- "page_id": "phys_00001",
- "cer": 0.18037974683544303,
- "wer": 0.1896551724137931
+ "page_id": "P_1877_7_0059",
+ "cer": 1.0,
+ "wer": 1.0
},
{
- "page_id": "phys_00002",
- "cer": 0.01878453038674033,
- "wer": 0.08670520231213873
+ "page_id": "P_1883_55_0044",
+ "cer": 1.0,
+ "wer": 1.0
},
{
- "page_id": "phys_00003",
- "cer": 0.02505446623093682,
- "wer": 0.08670520231213873
+ "page_id": "P_1929_250_0019",
+ "cer": 1.0,
+ "wer": 1.0
},
{
- "page_id": "phys_00004",
- "cer": 0.028397565922920892,
- "wer": 0.10344827586206896
+ "page_id": "P_1932_300_0488",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1936_123_0292",
+ "cer": 1.0,
+ "wer": 1.0
}
]
}
},
{
- "eval_workflow_id": "wf-databallenstedt_delatio_1777-eval",
- "label": "Workflow on data ballenstedt_delatio_1777",
+ "eval_workflow_id": "wf-datacalvi_beutelschneider01_1627-eval",
+ "label": "Workflow on data calvi_beutelschneider01_1627",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -1564,16 +2357,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/ballenstedt_delatio_1777.ocrd.zip",
- "label": "GT workspace ballenstedt_delatio_1777"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/calvi_beutelschneider01_1627.ocrd.zip",
+ "label": "GT workspace calvi_beutelschneider01_1627"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/ballenstedt_delatio_1777_ocr.zip",
- "label": "OCR workspace for ballenstedt_delatio_1777"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/calvi_beutelschneider01_1627_ocr.zip",
+ "label": "OCR workspace for calvi_beutelschneider01_1627"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/ballenstedt_delatio_1777_evaluation.zip",
- "label": "Evaluation workspace for ballenstedt_delatio_1777"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/calvi_beutelschneider01_1627_evaluation.zip",
+ "label": "Evaluation workspace for calvi_beutelschneider01_1627"
},
"workflow_steps": [
{
@@ -1611,9 +2404,9 @@
"Antiqua",
"Fraktur"
],
- "publication_century": "1700-1800",
+ "publication_century": "1600-1700",
"publication_decade": "",
- "publication_year": "18th century",
+ "publication_year": "17th century",
"number_of_pages": 3,
"layout": "simple"
}
@@ -1621,75 +2414,155 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 21.444519,
- "cpu_time": 24.028484,
- "cer_mean": 0.06438361522903834,
- "cer_median": 0.03969957081545064,
+ "wall_time": 6.47648,
+ "cpu_time": 8.9799,
+ "cer_mean": 0.08254540344028655,
+ "cer_median": 0.09331797235023041,
"cer_range": [
- 0.021764032073310423,
- 0.13168724279835392
+ 0.05025996533795494,
+ 0.1040582726326743
],
- "cer_standard_deviation": 0.058972490200809365,
- "wer": 0.16906902212925057,
- "pages_per_minute": 8.393753200992757
+ "cer_standard_deviation": 0.02847104928930229,
+ "wer": 0.1821206964037463,
+ "pages_per_minute": 27.792875142052477
},
"by_page": [
{
- "page_id": "phys_00003",
- "cer": 0.021764032073310423,
- "wer": 0.12236286919831224
+ "page_id": "phys_0001",
+ "cer": 0.1040582726326743,
+ "wer": 0.21656050955414013
},
{
- "page_id": "phys_00005",
- "cer": 0.13168724279835392,
- "wer": 0.2603305785123967
+ "page_id": "phys_0002",
+ "cer": 0.09331797235023041,
+ "wer": 0.20689655172413793
},
{
- "page_id": "phys_00010",
- "cer": 0.03969957081545064,
- "wer": 0.1245136186770428
+ "page_id": "phys_0003",
+ "cer": 0.05025996533795494,
+ "wer": 0.12290502793296089
}
]
}
},
{
- "eval_workflow_id": "wf-datareichsanzeiger_many_ads-eval",
- "label": "Workflow on data reichsanzeiger_many_ads",
+ "eval_workflow_id": "wf-dataweigel_gnothi02_1618-eval",
+ "label": "Workflow on data weigel_gnothi02_1618",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_many_ads.ocrd.zip",
- "label": "GT workspace reichsanzeiger_many_ads"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/weigel_gnothi02_1618.ocrd.zip",
+ "label": "GT workspace weigel_gnothi02_1618"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_many_ads_ocr.zip",
- "label": "OCR workspace for reichsanzeiger_many_ads"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/weigel_gnothi02_1618_ocr.zip",
+ "label": "OCR workspace for weigel_gnothi02_1618"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_many_ads_evaluation.zip",
- "label": "Evaluation workspace for reichsanzeiger_many_ads"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/weigel_gnothi02_1618_evaluation.zip",
+ "label": "Evaluation workspace for weigel_gnothi02_1618"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
"find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "overwrite_segments": false,
+ "segmentation_level": "word",
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -1709,113 +2582,8362 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua"
+ "Antiqua",
+ "Fraktur"
],
- "publication_century": "1820-1939",
+ "publication_century": "1600-1700",
"publication_decade": "",
- "publication_year": "19th century",
- "number_of_pages": 5,
- "layout": "reichsanzeiger-gt"
+ "publication_year": "17th century",
+ "number_of_pages": 4,
+ "layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 344.194331,
- "cpu_time": 401.989378,
- "cer_mean": 1.3340240550306781,
- "cer_median": 0.7412443874278384,
+ "wall_time": 73.18760400000001,
+ "cpu_time": 107.389301,
+ "cer_mean": 0.07997168401781861,
+ "cer_median": 0.07157755385146221,
"cer_range": [
- 0.3960932753867003,
- 3.7402255639097746
+ 0.043933054393305436,
+ 0.13279857397504458
],
- "cer_standard_deviation": 1.378192503303974,
- "wer": 1.8653811990174032,
- "pages_per_minute": 0.8716006423708356
+ "cer_standard_deviation": 0.03821904307003693,
+ "wer": 0.18620716224746828,
+ "pages_per_minute": 3.2792438457201025
},
"by_page": [
{
- "page_id": "P_1871_155_0279",
- "cer": 0.3960932753867003,
- "wer": 0.49564980967917344
- },
- {
- "page_id": "P_1871_65_0045",
- "cer": 0.7412443874278384,
- "wer": 0.947814451382694
+ "page_id": "phys_0001",
+ "cer": 0.13279857397504458,
+ "wer": 0.2717948717948718
},
{
- "page_id": "P_1873_1_0017",
- "cer": 0.5852251348300515,
- "wer": 0.6788418708240535
+ "page_id": "phys_0002",
+ "cer": 0.06286836935166994,
+ "wer": 0.20948616600790515
},
{
- "page_id": "P_1881_1_0662",
- "cer": 3.7402255639097746,
- "wer": 5.734011627906977
+ "page_id": "phys_0003",
+ "cer": 0.08028673835125448,
+ "wer": 0.16483516483516483
},
{
- "page_id": "P_1883_55_0044",
- "cer": 1.2073319135990264,
- "wer": 1.4705882352941178
+ "page_id": "phys_0004",
+ "cer": 0.043933054393305436,
+ "wer": 0.09871244635193133
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-databenner_herrnhuterey04_1748-eval",
+ "label": "Workflow on data benner_herrnhuterey04_1748",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/benner_herrnhuterey04_1748.ocrd.zip",
+ "label": "GT workspace benner_herrnhuterey04_1748"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/benner_herrnhuterey04_1748_ocr.zip",
+ "label": "OCR workspace for benner_herrnhuterey04_1748"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/benner_herrnhuterey04_1748_evaluation.zip",
+ "label": "Evaluation workspace for benner_herrnhuterey04_1748"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 17.756173,
+ "cpu_time": 21.102642,
+ "cer_mean": 0.20433470587543373,
+ "cer_median": 0.1700173533179325,
+ "cer_range": [
+ 0.04063745019920319,
+ 0.43666666666666665
+ ],
+ "cer_standard_deviation": 0.19097881579976753,
+ "wer": 0.31449770246107817,
+ "pages_per_minute": 13.51642609023915
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.43666666666666665,
+ "wer": 0.583941605839416
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.04063745019920319,
+ "wer": 0.10344827586206896
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.05536912751677853,
+ "wer": 0.18435754189944134
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.28466557911908646,
+ "wer": 0.3862433862433862
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datacalvi_beutelschneider01_1627-eval",
+ "label": "Workflow on data calvi_beutelschneider01_1627",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/calvi_beutelschneider01_1627.ocrd.zip",
+ "label": "GT workspace calvi_beutelschneider01_1627"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/calvi_beutelschneider01_1627_ocr.zip",
+ "label": "OCR workspace for calvi_beutelschneider01_1627"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/calvi_beutelschneider01_1627_evaluation.zip",
+ "label": "Evaluation workspace for calvi_beutelschneider01_1627"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 133,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 27.008717,
+ "cpu_time": 26.946455000000004,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 6.664515015652168
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataalberti_pictura_1540-eval",
+ "label": "Workflow on data alberti_pictura_1540",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/alberti_pictura_1540.ocrd.zip",
+ "label": "GT workspace alberti_pictura_1540"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/alberti_pictura_1540_ocr.zip",
+ "label": "OCR workspace for alberti_pictura_1540"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/alberti_pictura_1540_evaluation.zip",
+ "label": "Evaluation workspace for alberti_pictura_1540"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 3,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 14.196548,
+ "cpu_time": 13.619762999999999,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 12.67913861876845
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0007",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0008",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0009",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datakistler_kraeuter_1500-eval",
+ "label": "Workflow on data kistler_kraeuter_1500",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/kistler_kraeuter_1500.ocrd.zip",
+ "label": "GT workspace kistler_kraeuter_1500"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/kistler_kraeuter_1500_ocr.zip",
+ "label": "OCR workspace for kistler_kraeuter_1500"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/kistler_kraeuter_1500_evaluation.zip",
+ "label": "Evaluation workspace for kistler_kraeuter_1500"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 7.076052,
+ "cpu_time": 8.821135,
+ "cer_mean": 0.1809783660358373,
+ "cer_median": 0.1809783660358373,
+ "cer_range": [
+ 0.13479623824451412,
+ 0.2271604938271605
+ ],
+ "cer_standard_deviation": 0.06531139146173669,
+ "wer": 0.408373786407767,
+ "pages_per_minute": 16.95860912271419
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0007",
+ "cer": 0.2271604938271605,
+ "wer": 0.441747572815534
+ },
+ {
+ "page_id": "phys_0021",
+ "cer": 0.13479623824451412,
+ "wer": 0.375
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datareichsanzeiger_tables-eval",
+ "label": "Workflow on data reichsanzeiger_tables",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_tables.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_tables"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_tables_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_tables"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_tables_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_tables"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua"
+ ],
+ "publication_century": "1820-1939",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 5,
+ "layout": "reichsanzeiger-gt"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 437.353829,
+ "cpu_time": 575.346313,
+ "cer_mean": 1.479972651687655,
+ "cer_median": 1.331401349741961,
+ "cer_range": [
+ 1.090097148229395,
+ 1.9778823058446757
+ ],
+ "cer_standard_deviation": 0.3854879135899152,
+ "wer": 1.7910299442503619,
+ "pages_per_minute": 0.6859434629529676
+ },
+ "by_page": [
+ {
+ "page_id": "P_1877_7_0059",
+ "cer": 1.090097148229395,
+ "wer": 1.4218399401645474
+ },
+ {
+ "page_id": "P_1883_55_0044",
+ "cer": 1.2073319135990264,
+ "wer": 1.4705882352941178
+ },
+ {
+ "page_id": "P_1929_250_0019",
+ "cer": 1.331401349741961,
+ "wer": 1.7407221664994985
+ },
+ {
+ "page_id": "P_1932_300_0488",
+ "cer": 1.9778823058446757,
+ "wer": 2.349493487698987
+ },
+ {
+ "page_id": "P_1936_123_0292",
+ "cer": 1.7931505410232167,
+ "wer": 1.9725058915946583
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datareichsanzeiger_random-eval",
+ "label": "Workflow on data reichsanzeiger_random",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_random.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_random"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_random_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_random"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_random_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_random"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua"
+ ],
+ "publication_century": "1820-1939",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 6,
+ "layout": "reichsanzeiger-gt"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 577.257552,
+ "cpu_time": 763.98953,
+ "cer_mean": 1.1094634128044334,
+ "cer_median": 0.8993673021608106,
+ "cer_range": [
+ 0.30019453260980855,
+ 1.9778823058446757
+ ],
+ "cer_standard_deviation": 0.645494993792005,
+ "wer": 1.3304151477151094,
+ "pages_per_minute": 0.6236384413728727
+ },
+ "by_page": [
+ {
+ "page_id": "P_1879_45_0344",
+ "cer": 0.9843462873477166,
+ "wer": 1.1759884281581485
+ },
+ {
+ "page_id": "P_1885_5_0055",
+ "cer": 0.30019453260980855,
+ "wer": 0.40127817019845274
+ },
+ {
+ "page_id": "P_1889_1_0018",
+ "cer": 0.7868184930272782,
+ "wer": 0.9704142011834319
+ },
+ {
+ "page_id": "P_1891_33_0452",
+ "cer": 0.8143883169739047,
+ "wer": 1.112810707456979
+ },
+ {
+ "page_id": "P_1932_300_0488",
+ "cer": 1.9778823058446757,
+ "wer": 2.349493487698987
+ },
+ {
+ "page_id": "P_1936_123_0292",
+ "cer": 1.7931505410232167,
+ "wer": 1.9725058915946583
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datarollenhagen_reysen_1603-eval",
+ "label": "Workflow on data rollenhagen_reysen_1603",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/rollenhagen_reysen_1603.ocrd.zip",
+ "label": "GT workspace rollenhagen_reysen_1603"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/rollenhagen_reysen_1603_ocr.zip",
+ "label": "OCR workspace for rollenhagen_reysen_1603"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/rollenhagen_reysen_1603_evaluation.zip",
+ "label": "Evaluation workspace for rollenhagen_reysen_1603"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 60.776995,
+ "cpu_time": 83.81908,
+ "cer_mean": 0.20072324583561607,
+ "cer_median": 0.19980314960629922,
+ "cer_range": [
+ 0.19198790627362056,
+ 0.21037868162692847
+ ],
+ "cer_standard_deviation": 0.009229847632943988,
+ "wer": 0.34944871614736,
+ "pages_per_minute": 2.9616469192002666
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.21037868162692847,
+ "wer": 0.4298245614035088
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.19980314960629922,
+ "wer": 0.3790849673202614
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.19198790627362056,
+ "wer": 0.23943661971830985
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datajusti_abhandlung01_1758-eval",
+ "label": "Workflow on data justi_abhandlung01_1758",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/justi_abhandlung01_1758.ocrd.zip",
+ "label": "GT workspace justi_abhandlung01_1758"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/justi_abhandlung01_1758_ocr.zip",
+ "label": "OCR workspace for justi_abhandlung01_1758"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/justi_abhandlung01_1758_evaluation.zip",
+ "label": "Evaluation workspace for justi_abhandlung01_1758"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 12.158114,
+ "cpu_time": 15.372055,
+ "cer_mean": 0.12942563348778183,
+ "cer_median": 0.13278777025964683,
+ "cer_range": [
+ 0.09893550407013149,
+ 0.15319148936170213
+ ],
+ "cer_standard_deviation": 0.022501323128153037,
+ "wer": 0.19483806092245076,
+ "pages_per_minute": 19.73990373835942
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.13375130616509928,
+ "wer": 0.2
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.15319148936170213,
+ "wer": 0.22916666666666666
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.09893550407013149,
+ "wer": 0.12992125984251968
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.1318242343541944,
+ "wer": 0.22026431718061673
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-databuerger_gedichte_1778-eval",
+ "label": "Workflow on data buerger_gedichte_1778",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/buerger_gedichte_1778.ocrd.zip",
+ "label": "GT workspace buerger_gedichte_1778"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/buerger_gedichte_1778_ocr.zip",
+ "label": "OCR workspace for buerger_gedichte_1778"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/buerger_gedichte_1778_evaluation.zip",
+ "label": "Evaluation workspace for buerger_gedichte_1778"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 2,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 15.012823999999998,
+ "cpu_time": 23.767021000000003,
+ "cer_mean": 0.22786023044476886,
+ "cer_median": 0.22786023044476886,
+ "cer_range": [
+ 0.053811659192825115,
+ 0.4019088016967126
+ ],
+ "cer_standard_deviation": 0.24614184997615882,
+ "wer": 0.35125551082997897,
+ "pages_per_minute": 7.993166375626599
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.4019088016967126,
+ "wer": 0.5673758865248227
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.053811659192825115,
+ "wer": 0.13513513513513514
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datatrota_mordtbrenner_1540-eval",
+ "label": "Workflow on data trota_mordtbrenner_1540",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/trota_mordtbrenner_1540.ocrd.zip",
+ "label": "GT workspace trota_mordtbrenner_1540"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/trota_mordtbrenner_1540_ocr.zip",
+ "label": "OCR workspace for trota_mordtbrenner_1540"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/trota_mordtbrenner_1540_evaluation.zip",
+ "label": "Evaluation workspace for trota_mordtbrenner_1540"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 16.12696,
+ "cpu_time": 15.415920999999999,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 7.440956014028682
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datasilesius_seelenlust01_1657-eval",
+ "label": "Workflow on data silesius_seelenlust01_1657",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/silesius_seelenlust01_1657.ocrd.zip",
+ "label": "GT workspace silesius_seelenlust01_1657"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/silesius_seelenlust01_1657_ocr.zip",
+ "label": "OCR workspace for silesius_seelenlust01_1657"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/silesius_seelenlust01_1657_evaluation.zip",
+ "label": "Evaluation workspace for silesius_seelenlust01_1657"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 5,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 48.011177999999994,
+ "cpu_time": 74.53987599999999,
+ "cer_mean": 0.16456973001701142,
+ "cer_median": 0.11229135053110774,
+ "cer_range": [
+ 0.05917159763313609,
+ 0.34146341463414637
+ ],
+ "cer_standard_deviation": 0.11554307430913685,
+ "wer": 0.34312458192490203,
+ "pages_per_minute": 6.248544870113373
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.05917159763313609,
+ "wer": 0.2463768115942029
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.0913884007029877,
+ "wer": 0.28125
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.34146341463414637,
+ "wer": 0.5774647887323944
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.11229135053110774,
+ "wer": 0.1693548387096774
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 0.21853388658367912,
+ "wer": 0.4411764705882353
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datalessing_menschengeschlecht_1780-eval",
+ "label": "Workflow on data lessing_menschengeschlecht_1780",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/lessing_menschengeschlecht_1780.ocrd.zip",
+ "label": "GT workspace lessing_menschengeschlecht_1780"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/lessing_menschengeschlecht_1780_ocr.zip",
+ "label": "OCR workspace for lessing_menschengeschlecht_1780"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/lessing_menschengeschlecht_1780_evaluation.zip",
+ "label": "Evaluation workspace for lessing_menschengeschlecht_1780"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 1,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 5.051292,
+ "cpu_time": 4.899471,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": null,
+ "wer": 1.0,
+ "pages_per_minute": 11.87814919430514
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataweigel_gnothi02_1618-eval",
+ "label": "Workflow on data weigel_gnothi02_1618",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/weigel_gnothi02_1618.ocrd.zip",
+ "label": "GT workspace weigel_gnothi02_1618"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/weigel_gnothi02_1618_ocr.zip",
+ "label": "OCR workspace for weigel_gnothi02_1618"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/weigel_gnothi02_1618_evaluation.zip",
+ "label": "Evaluation workspace for weigel_gnothi02_1618"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 4,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 25.781608,
+ "cpu_time": 29.469248,
+ "cer_mean": 0.11399763421865425,
+ "cer_median": 0.09987627660053089,
+ "cer_range": [
+ 0.07531106745252128,
+ 0.18092691622103388
+ ],
+ "cer_standard_deviation": 0.047514032403792344,
+ "wer": 0.19377061794956923,
+ "pages_per_minute": 9.308961644285338
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.18092691622103388,
+ "wer": 0.3282051282051282
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.07531106745252128,
+ "wer": 0.15019762845849802
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.11397849462365592,
+ "wer": 0.15934065934065933
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.08577405857740586,
+ "wer": 0.13733905579399142
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataestor_rechtsgelehrsamkeit02_1758-eval",
+ "label": "Workflow on data estor_rechtsgelehrsamkeit02_1758",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/estor_rechtsgelehrsamkeit02_1758.ocrd.zip",
+ "label": "GT workspace estor_rechtsgelehrsamkeit02_1758"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/estor_rechtsgelehrsamkeit02_1758_ocr.zip",
+ "label": "OCR workspace for estor_rechtsgelehrsamkeit02_1758"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/estor_rechtsgelehrsamkeit02_1758_evaluation.zip",
+ "label": "Evaluation workspace for estor_rechtsgelehrsamkeit02_1758"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 17.466218,
+ "cpu_time": 20.755792,
+ "cer_mean": 0.11589408928020027,
+ "cer_median": 0.10084584323499293,
+ "cer_range": [
+ 0.05331088664421998,
+ 0.2085737840065952
+ ],
+ "cer_standard_deviation": 0.07259145757108061,
+ "wer": 0.20102650242627845,
+ "pages_per_minute": 13.740810975793385
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.06310013717421124,
+ "wer": 0.11403508771929824
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.13859154929577464,
+ "wer": 0.22340425531914893
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.05331088664421998,
+ "wer": 0.13333333333333333
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.2085737840065952,
+ "wer": 0.3333333333333333
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataheyden_paedono_1548-eval",
+ "label": "Workflow on data heyden_paedono_1548",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/heyden_paedono_1548.ocrd.zip",
+ "label": "GT workspace heyden_paedono_1548"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/heyden_paedono_1548_ocr.zip",
+ "label": "OCR workspace for heyden_paedono_1548"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/heyden_paedono_1548_evaluation.zip",
+ "label": "Evaluation workspace for heyden_paedono_1548"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 7.339199,
+ "cpu_time": 9.278232,
+ "cer_mean": 0.07452119312897007,
+ "cer_median": 0.0629800307219662,
+ "cer_range": [
+ 0.037037037037037035,
+ 0.12354651162790697
+ ],
+ "cer_standard_deviation": 0.044394494261965886,
+ "wer": 0.22683890077340793,
+ "pages_per_minute": 24.525837220110805
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0007",
+ "cer": 0.0629800307219662,
+ "wer": 0.21359223300970873
+ },
+ {
+ "page_id": "phys_0013",
+ "cer": 0.12354651162790697,
+ "wer": 0.33980582524271846
+ },
+ {
+ "page_id": "phys_0014",
+ "cer": 0.037037037037037035,
+ "wer": 0.1271186440677966
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-databuerger_gedichte_1778-eval",
+ "label": "Workflow on data buerger_gedichte_1778",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/buerger_gedichte_1778.ocrd.zip",
+ "label": "GT workspace buerger_gedichte_1778"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/buerger_gedichte_1778_ocr.zip",
+ "label": "OCR workspace for buerger_gedichte_1778"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/buerger_gedichte_1778_evaluation.zip",
+ "label": "Evaluation workspace for buerger_gedichte_1778"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 2,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 5.502369,
+ "cpu_time": 6.645983,
+ "cer_mean": 0.10672693293515115,
+ "cer_median": 0.10672693293515115,
+ "cer_range": [
+ 0.04484304932735426,
+ 0.16861081654294804
+ ],
+ "cer_standard_deviation": 0.08751702749046443,
+ "wer": 0.23135901859306116,
+ "pages_per_minute": 21.80878817832828
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.16861081654294804,
+ "wer": 0.3546099290780142
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.04484304932735426,
+ "wer": 0.10810810810810811
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataluther_auszlegunge_1520-eval",
+ "label": "Workflow on data luther_auszlegunge_1520",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/luther_auszlegunge_1520.ocrd.zip",
+ "label": "GT workspace luther_auszlegunge_1520"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luther_auszlegunge_1520_ocr.zip",
+ "label": "OCR workspace for luther_auszlegunge_1520"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luther_auszlegunge_1520_evaluation.zip",
+ "label": "Evaluation workspace for luther_auszlegunge_1520"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 18.242696,
+ "cpu_time": 17.645239,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 6.577975097540407
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0029",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-databohse_helicon_1696-eval",
+ "label": "Workflow on data bohse_helicon_1696",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/bohse_helicon_1696.ocrd.zip",
+ "label": "GT workspace bohse_helicon_1696"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bohse_helicon_1696_ocr.zip",
+ "label": "OCR workspace for bohse_helicon_1696"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bohse_helicon_1696_evaluation.zip",
+ "label": "Evaluation workspace for bohse_helicon_1696"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 5,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 51.137511999999994,
+ "cpu_time": 78.57060399999999,
+ "cer_mean": 0.17535728446779447,
+ "cer_median": 0.14864864864864866,
+ "cer_range": [
+ 0.08232118758434548,
+ 0.35185185185185186
+ ],
+ "cer_standard_deviation": 0.1049688941973522,
+ "wer": 0.3836611930189032,
+ "pages_per_minute": 5.866534922543749
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.35185185185185186,
+ "wer": 0.7021276595744681
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.14864864864864866,
+ "wer": 0.35185185185185186
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.1781609195402299,
+ "wer": 0.381294964028777
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.08232118758434548,
+ "wer": 0.24074074074074073
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 0.11580381471389646,
+ "wer": 0.2422907488986784
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datalessing_menschengeschlecht_1780-eval",
+ "label": "Workflow on data lessing_menschengeschlecht_1780",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/lessing_menschengeschlecht_1780.ocrd.zip",
+ "label": "GT workspace lessing_menschengeschlecht_1780"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/lessing_menschengeschlecht_1780_ocr.zip",
+ "label": "OCR workspace for lessing_menschengeschlecht_1780"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/lessing_menschengeschlecht_1780_evaluation.zip",
+ "label": "Evaluation workspace for lessing_menschengeschlecht_1780"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 1,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 7.084619,
+ "cpu_time": 10.402543999999999,
+ "cer_mean": 0.0199501246882793,
+ "cer_median": 0.0199501246882793,
+ "cer_range": [
+ 0.0199501246882793,
+ 0.0199501246882793
+ ],
+ "cer_standard_deviation": null,
+ "wer": 0.09836065573770492,
+ "pages_per_minute": 8.469051052710103
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.0199501246882793,
+ "wer": 0.09836065573770492
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datareichsanzeiger_random-eval",
+ "label": "Workflow on data reichsanzeiger_random",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_random.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_random"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_random_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_random"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_random_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_random"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-crop",
+ "params": {
+ "dpi": 0,
+ "padding": 4
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua"
+ ],
+ "publication_century": "1820-1939",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 6,
+ "layout": "reichsanzeiger-gt"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 1758.9687809999998,
+ "cpu_time": 1685.08457,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 0.20466537205699278
+ },
+ "by_page": [
+ {
+ "page_id": "P_1879_45_0344",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1885_5_0055",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1889_1_0018",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1891_33_0452",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1932_300_0488",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1936_123_0292",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datasilesius_seelenlust01_1657-eval",
+ "label": "Workflow on data silesius_seelenlust01_1657",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/silesius_seelenlust01_1657.ocrd.zip",
+ "label": "GT workspace silesius_seelenlust01_1657"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/silesius_seelenlust01_1657_ocr.zip",
+ "label": "OCR workspace for silesius_seelenlust01_1657"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/silesius_seelenlust01_1657_evaluation.zip",
+ "label": "Evaluation workspace for silesius_seelenlust01_1657"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 5,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 11.083143,
+ "cpu_time": 14.729302,
+ "cer_mean": 0.3007055286105995,
+ "cer_median": 0.2951219512195122,
+ "cer_range": [
+ 0.19271623672230653,
+ 0.44970414201183434
+ ],
+ "cer_standard_deviation": 0.10657123719012947,
+ "wer": 0.5174305966287508,
+ "pages_per_minute": 27.068134012166045
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.44970414201183434,
+ "wer": 0.7101449275362319
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.20913884007029876,
+ "wer": 0.5104166666666666
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.2951219512195122,
+ "wer": 0.4647887323943662
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.19271623672230653,
+ "wer": 0.3870967741935484
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 0.35684647302904565,
+ "wer": 0.5147058823529411
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datasilesius_seelenlust01_1657-eval",
+ "label": "Workflow on data silesius_seelenlust01_1657",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/silesius_seelenlust01_1657.ocrd.zip",
+ "label": "GT workspace silesius_seelenlust01_1657"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/silesius_seelenlust01_1657_ocr.zip",
+ "label": "OCR workspace for silesius_seelenlust01_1657"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/silesius_seelenlust01_1657_evaluation.zip",
+ "label": "Evaluation workspace for silesius_seelenlust01_1657"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 5,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 25.224073,
+ "cpu_time": 24.197834,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 11.89340040365408
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataheyden_paedono_1548-eval",
+ "label": "Workflow on data heyden_paedono_1548",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/heyden_paedono_1548.ocrd.zip",
+ "label": "GT workspace heyden_paedono_1548"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/heyden_paedono_1548_ocr.zip",
+ "label": "OCR workspace for heyden_paedono_1548"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/heyden_paedono_1548_evaluation.zip",
+ "label": "Evaluation workspace for heyden_paedono_1548"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 23.972654,
+ "cpu_time": 38.630559,
+ "cer_mean": 0.06721136853878373,
+ "cer_median": 0.055232558139534885,
+ "cer_range": [
+ 0.03580246913580247,
+ 0.11059907834101383
+ ],
+ "cer_standard_deviation": 0.038810463938030185,
+ "wer": 0.1847677033624047,
+ "pages_per_minute": 7.508555373134739
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0007",
+ "cer": 0.11059907834101383,
+ "wer": 0.23300970873786409
+ },
+ {
+ "page_id": "phys_0013",
+ "cer": 0.055232558139534885,
+ "wer": 0.1941747572815534
+ },
+ {
+ "page_id": "phys_0014",
+ "cer": 0.03580246913580247,
+ "wer": 0.1271186440677966
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataestor_rechtsgelehrsamkeit02_1758-eval",
+ "label": "Workflow on data estor_rechtsgelehrsamkeit02_1758",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/estor_rechtsgelehrsamkeit02_1758.ocrd.zip",
+ "label": "GT workspace estor_rechtsgelehrsamkeit02_1758"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/estor_rechtsgelehrsamkeit02_1758_ocr.zip",
+ "label": "OCR workspace for estor_rechtsgelehrsamkeit02_1758"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/estor_rechtsgelehrsamkeit02_1758_evaluation.zip",
+ "label": "Evaluation workspace for estor_rechtsgelehrsamkeit02_1758"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 53.64139899999999,
+ "cpu_time": 84.639989,
+ "cer_mean": 0.2886254685112176,
+ "cer_median": 0.2965426192770805,
+ "cer_range": [
+ 0.07856341189674523,
+ 0.4828532235939643
+ ],
+ "cer_standard_deviation": 0.204671750373206,
+ "wer": 0.4238352166683633,
+ "pages_per_minute": 4.474156238915395
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.4828532235939643,
+ "wer": 0.5614035087719298
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.14873239436619717,
+ "wer": 0.2695035460992908
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.07856341189674523,
+ "wer": 0.22807017543859648
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.4443528441879637,
+ "wer": 0.6363636363636364
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datalessing_menschengeschlecht_1780-eval",
+ "label": "Workflow on data lessing_menschengeschlecht_1780",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/lessing_menschengeschlecht_1780.ocrd.zip",
+ "label": "GT workspace lessing_menschengeschlecht_1780"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/lessing_menschengeschlecht_1780_ocr.zip",
+ "label": "OCR workspace for lessing_menschengeschlecht_1780"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/lessing_menschengeschlecht_1780_evaluation.zip",
+ "label": "Evaluation workspace for lessing_menschengeschlecht_1780"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 1,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 1.7175,
+ "cpu_time": 2.152713,
+ "cer_mean": 0.02493765586034913,
+ "cer_median": 0.02493765586034913,
+ "cer_range": [
+ 0.02493765586034913,
+ 0.02493765586034913
+ ],
+ "cer_standard_deviation": null,
+ "wer": 0.09836065573770492,
+ "pages_per_minute": 34.93449781659388
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.02493765586034913,
+ "wer": 0.09836065573770492
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataeuler_rechenkunst01_1738-eval",
+ "label": "Workflow on data euler_rechenkunst01_1738",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/euler_rechenkunst01_1738.ocrd.zip",
+ "label": "GT workspace euler_rechenkunst01_1738"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/euler_rechenkunst01_1738_ocr.zip",
+ "label": "OCR workspace for euler_rechenkunst01_1738"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/euler_rechenkunst01_1738_evaluation.zip",
+ "label": "Evaluation workspace for euler_rechenkunst01_1738"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 201,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 6,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 44.243076,
+ "cpu_time": 43.668200000000006,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 8.136866433066272
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0006",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-databuerger_gedichte_1778-eval",
+ "label": "Workflow on data buerger_gedichte_1778",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/buerger_gedichte_1778.ocrd.zip",
+ "label": "GT workspace buerger_gedichte_1778"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/buerger_gedichte_1778_ocr.zip",
+ "label": "OCR workspace for buerger_gedichte_1778"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/buerger_gedichte_1778_evaluation.zip",
+ "label": "Evaluation workspace for buerger_gedichte_1778"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 2,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 9.932442,
+ "cpu_time": 9.373615,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 12.081621015254857
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datann_besuch_1780-eval",
+ "label": "Workflow on data nn_besuch_1780",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/nn_besuch_1780.ocrd.zip",
+ "label": "GT workspace nn_besuch_1780"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/nn_besuch_1780_ocr.zip",
+ "label": "OCR workspace for nn_besuch_1780"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/nn_besuch_1780_evaluation.zip",
+ "label": "Evaluation workspace for nn_besuch_1780"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 30.565247,
+ "cpu_time": 29.36571,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 7.852054982575472
+ },
+ "by_page": [
+ {
+ "page_id": "phys_00001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_00002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_00003",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_00004",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataluther_auszlegunge_1520-eval",
+ "label": "Workflow on data luther_auszlegunge_1520",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/luther_auszlegunge_1520.ocrd.zip",
+ "label": "GT workspace luther_auszlegunge_1520"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luther_auszlegunge_1520_ocr.zip",
+ "label": "OCR workspace for luther_auszlegunge_1520"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luther_auszlegunge_1520_evaluation.zip",
+ "label": "Evaluation workspace for luther_auszlegunge_1520"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 9.141584,
+ "cpu_time": 10.972897,
+ "cer_mean": 0.24855845660550213,
+ "cer_median": 0.24855845660550213,
+ "cer_range": [
+ 0.07865168539325842,
+ 0.4184652278177458
+ ],
+ "cer_standard_deviation": 0.2402844601873776,
+ "wer": 0.37300307219662054,
+ "pages_per_minute": 13.126827910786577
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0003",
+ "cer": 0.4184652278177458,
+ "wer": 0.48333333333333334
+ },
+ {
+ "page_id": "phys_0029",
+ "cer": 0.07865168539325842,
+ "wer": 0.2626728110599078
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datatrota_mordtbrenner_1540-eval",
+ "label": "Workflow on data trota_mordtbrenner_1540",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/trota_mordtbrenner_1540.ocrd.zip",
+ "label": "GT workspace trota_mordtbrenner_1540"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/trota_mordtbrenner_1540_ocr.zip",
+ "label": "OCR workspace for trota_mordtbrenner_1540"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/trota_mordtbrenner_1540_evaluation.zip",
+ "label": "Evaluation workspace for trota_mordtbrenner_1540"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 24.812708,
+ "cpu_time": 34.840120999999996,
+ "cer_mean": 0.07499762380001901,
+ "cer_median": 0.07499762380001901,
+ "cer_range": [
+ 0.0658682634730539,
+ 0.08412698412698413
+ ],
+ "cer_standard_deviation": 0.012910865190184943,
+ "wer": 0.15174388339406558,
+ "pages_per_minute": 4.836231498794891
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.0658682634730539,
+ "wer": 0.18584070796460178
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.08412698412698413,
+ "wer": 0.11764705882352941
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-databenner_herrnhuterey04_1748-eval",
+ "label": "Workflow on data benner_herrnhuterey04_1748",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/benner_herrnhuterey04_1748.ocrd.zip",
+ "label": "GT workspace benner_herrnhuterey04_1748"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/benner_herrnhuterey04_1748_ocr.zip",
+ "label": "OCR workspace for benner_herrnhuterey04_1748"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/benner_herrnhuterey04_1748_evaluation.zip",
+ "label": "Evaluation workspace for benner_herrnhuterey04_1748"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 37.143108999999995,
+ "cpu_time": 35.83222599999999,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 6.46149464763437
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datablumenbach_anatomie_1805-eval",
+ "label": "Workflow on data blumenbach_anatomie_1805",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/blumenbach_anatomie_1805.ocrd.zip",
+ "label": "GT workspace blumenbach_anatomie_1805"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/blumenbach_anatomie_1805_ocr.zip",
+ "label": "OCR workspace for blumenbach_anatomie_1805"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/blumenbach_anatomie_1805_evaluation.zip",
+ "label": "Evaluation workspace for blumenbach_anatomie_1805"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1800-1900",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 11.765522,
+ "cpu_time": 14.113986,
+ "cer_mean": 0.08328200324172261,
+ "cer_median": 0.08736842105263158,
+ "cer_range": [
+ 0.04055496264674493,
+ 0.12192262602579132
+ ],
+ "cer_standard_deviation": 0.04083746158658049,
+ "wer": 0.23519468186134854,
+ "pages_per_minute": 15.298938712621506
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.08736842105263158,
+ "wer": 0.22666666666666666
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.04055496264674493,
+ "wer": 0.14814814814814814
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.12192262602579132,
+ "wer": 0.33076923076923076
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datarollenhagen_reysen_1603-eval",
+ "label": "Workflow on data rollenhagen_reysen_1603",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/rollenhagen_reysen_1603.ocrd.zip",
+ "label": "GT workspace rollenhagen_reysen_1603"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/rollenhagen_reysen_1603_ocr.zip",
+ "label": "OCR workspace for rollenhagen_reysen_1603"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/rollenhagen_reysen_1603_evaluation.zip",
+ "label": "Evaluation workspace for rollenhagen_reysen_1603"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 14.252525,
+ "cpu_time": 16.514329,
+ "cer_mean": 0.21046220070684576,
+ "cer_median": 0.16929133858267717,
+ "cer_range": [
+ 0.14512471655328799,
+ 0.3169705469845722
+ ],
+ "cer_standard_deviation": 0.093027024434784,
+ "wer": 0.3190752126565147,
+ "pages_per_minute": 12.629341116749487
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.3169705469845722,
+ "wer": 0.4649122807017544
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.16929133858267717,
+ "wer": 0.28104575163398693
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.14512471655328799,
+ "wer": 0.2112676056338028
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-databohse_helicon_1696-eval",
+ "label": "Workflow on data bohse_helicon_1696",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/bohse_helicon_1696.ocrd.zip",
+ "label": "GT workspace bohse_helicon_1696"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bohse_helicon_1696_ocr.zip",
+ "label": "OCR workspace for bohse_helicon_1696"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bohse_helicon_1696_evaluation.zip",
+ "label": "Evaluation workspace for bohse_helicon_1696"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 315,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 5,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 28.21906,
+ "cpu_time": 27.394986,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 10.63111244669383
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataarnimb_goethe03_1835-eval",
+ "label": "Workflow on data arnimb_goethe03_1835",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/arnimb_goethe03_1835.ocrd.zip",
+ "label": "GT workspace arnimb_goethe03_1835"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/arnimb_goethe03_1835_ocr.zip",
+ "label": "OCR workspace for arnimb_goethe03_1835"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/arnimb_goethe03_1835_evaluation.zip",
+ "label": "Evaluation workspace for arnimb_goethe03_1835"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Fraktur"
+ ],
+ "publication_century": "1800-1900",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 1,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 4.205169,
+ "cpu_time": 5.02776,
+ "cer_mean": 0.004721435316336166,
+ "cer_median": 0.004721435316336166,
+ "cer_range": [
+ 0.004721435316336166,
+ 0.004721435316336166
+ ],
+ "cer_standard_deviation": null,
+ "wer": 0.015873015873015872,
+ "pages_per_minute": 14.268154264430276
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.004721435316336166,
+ "wer": 0.015873015873015872
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataluz_blitz_1784-eval",
+ "label": "Workflow on data luz_blitz_1784",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/luz_blitz_1784.ocrd.zip",
+ "label": "GT workspace luz_blitz_1784"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luz_blitz_1784_ocr.zip",
+ "label": "OCR workspace for luz_blitz_1784"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luz_blitz_1784_evaluation.zip",
+ "label": "Evaluation workspace for luz_blitz_1784"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 55.558143,
+ "cpu_time": 84.72784899999999,
+ "cer_mean": 0.02547899799369996,
+ "cer_median": 0.027204160076875337,
+ "cer_range": [
+ 0.010332950631458095,
+ 0.03717472118959108
+ ],
+ "cer_standard_deviation": 0.011266997500136374,
+ "wer": 0.07714749104131867,
+ "pages_per_minute": 4.319798809690237
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.03717472118959108,
+ "wer": 0.10256410256410256
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.025119617224880382,
+ "wer": 0.09565217391304348
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.029288702928870293,
+ "wer": 0.07924528301886792
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.010332950631458095,
+ "wer": 0.0311284046692607
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataheyden_paedono_1548-eval",
+ "label": "Workflow on data heyden_paedono_1548",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/heyden_paedono_1548.ocrd.zip",
+ "label": "GT workspace heyden_paedono_1548"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/heyden_paedono_1548_ocr.zip",
+ "label": "OCR workspace for heyden_paedono_1548"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/heyden_paedono_1548_evaluation.zip",
+ "label": "Evaluation workspace for heyden_paedono_1548"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 13.685353,
+ "cpu_time": 13.153721,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 13.15274805114636
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0007",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0013",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0014",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datarollenhagen_reysen_1603-eval",
+ "label": "Workflow on data rollenhagen_reysen_1603",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/rollenhagen_reysen_1603.ocrd.zip",
+ "label": "GT workspace rollenhagen_reysen_1603"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/rollenhagen_reysen_1603_ocr.zip",
+ "label": "OCR workspace for rollenhagen_reysen_1603"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/rollenhagen_reysen_1603_evaluation.zip",
+ "label": "Evaluation workspace for rollenhagen_reysen_1603"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 34.360988,
+ "cpu_time": 33.38291099999999,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 5.238498962835411
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-databallenstedt_delatio_1777-eval",
+ "label": "Workflow on data ballenstedt_delatio_1777",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/ballenstedt_delatio_1777.ocrd.zip",
+ "label": "GT workspace ballenstedt_delatio_1777"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/ballenstedt_delatio_1777_ocr.zip",
+ "label": "OCR workspace for ballenstedt_delatio_1777"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/ballenstedt_delatio_1777_evaluation.zip",
+ "label": "Evaluation workspace for ballenstedt_delatio_1777"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 56.39011899999999,
+ "cpu_time": 88.809237,
+ "cer_mean": 0.02808165896942696,
+ "cer_median": 0.02821869488536155,
+ "cer_range": [
+ 0.020618556701030927,
+ 0.03540772532188841
+ ],
+ "cer_standard_deviation": 0.007395536576593408,
+ "wer": 0.13838629881265765,
+ "pages_per_minute": 3.1920485927685314
+ },
+ "by_page": [
+ {
+ "page_id": "phys_00003",
+ "cer": 0.020618556701030927,
+ "wer": 0.11392405063291139
+ },
+ {
+ "page_id": "phys_00005",
+ "cer": 0.02821869488536155,
+ "wer": 0.16115702479338842
+ },
+ {
+ "page_id": "phys_00010",
+ "cer": 0.03540772532188841,
+ "wer": 0.14007782101167315
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataalberti_pictura_1540-eval",
+ "label": "Workflow on data alberti_pictura_1540",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/alberti_pictura_1540.ocrd.zip",
+ "label": "GT workspace alberti_pictura_1540"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/alberti_pictura_1540_ocr.zip",
+ "label": "OCR workspace for alberti_pictura_1540"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/alberti_pictura_1540_evaluation.zip",
+ "label": "Evaluation workspace for alberti_pictura_1540"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 3,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 27.849722,
+ "cpu_time": 47.006445,
+ "cer_mean": 0.10363204260868718,
+ "cer_median": 0.11836734693877551,
+ "cer_range": [
+ 0.046632124352331605,
+ 0.1458966565349544
+ ],
+ "cer_standard_deviation": 0.05124654849483992,
+ "wer": 0.2145458690579216,
+ "pages_per_minute": 6.463260207767963
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0007",
+ "cer": 0.046632124352331605,
+ "wer": 0.1652892561983471
+ },
+ {
+ "page_id": "phys_0008",
+ "cer": 0.1458966565349544,
+ "wer": 0.2670807453416149
+ },
+ {
+ "page_id": "phys_0009",
+ "cer": 0.11836734693877551,
+ "wer": 0.2112676056338028
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataluz_blitz_1784-eval",
+ "label": "Workflow on data luz_blitz_1784",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/luz_blitz_1784.ocrd.zip",
+ "label": "GT workspace luz_blitz_1784"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luz_blitz_1784_ocr.zip",
+ "label": "OCR workspace for luz_blitz_1784"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luz_blitz_1784_evaluation.zip",
+ "label": "Evaluation workspace for luz_blitz_1784"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 31.967221,
+ "cpu_time": 30.880905,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 7.50769045579533
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-databernd_lebensbeschreibung_1738-eval",
+ "label": "Workflow on data bernd_lebensbeschreibung_1738",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/bernd_lebensbeschreibung_1738.ocrd.zip",
+ "label": "GT workspace bernd_lebensbeschreibung_1738"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bernd_lebensbeschreibung_1738_ocr.zip",
+ "label": "OCR workspace for bernd_lebensbeschreibung_1738"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bernd_lebensbeschreibung_1738_evaluation.zip",
+ "label": "Evaluation workspace for bernd_lebensbeschreibung_1738"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 3,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 16.878657,
+ "cpu_time": 16.269429000000002,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 10.664355582319137
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datareichsanzeiger_random-eval",
+ "label": "Workflow on data reichsanzeiger_random",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_random.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_random"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_random_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_random"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_random_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_random"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-crop",
+ "params": {
+ "dpi": 0,
+ "padding": 4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua"
+ ],
+ "publication_century": "1820-1939",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 6,
+ "layout": "reichsanzeiger-gt"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 3106.4616029999997,
+ "cpu_time": 4679.73138,
+ "cer_mean": 1.0169660871444868,
+ "cer_median": 0.7215611935116096,
+ "cer_range": [
+ 0.18319367967440747,
+ 2.105984787830264
+ ],
+ "cer_standard_deviation": 0.8374705743292857,
+ "wer": 1.3445096706748612,
+ "pages_per_minute": 0.1158874777825477
+ },
+ "by_page": [
+ {
+ "page_id": "P_1879_45_0344",
+ "cer": 0.9179881576260804,
+ "wer": 1.1783992285438765
+ },
+ {
+ "page_id": "P_1885_5_0055",
+ "cer": 0.37306747209992835,
+ "wer": 0.5381769256643122
+ },
+ {
+ "page_id": "P_1889_1_0018",
+ "cer": 0.5251342293971388,
+ "wer": 0.8080473372781065
+ },
+ {
+ "page_id": "P_1891_33_0452",
+ "cer": 0.18319367967440747,
+ "wer": 0.4236050756127238
+ },
+ {
+ "page_id": "P_1932_300_0488",
+ "cer": 2.105984787830264,
+ "wer": 2.814037626628075
+ },
+ {
+ "page_id": "P_1936_123_0292",
+ "cer": 1.9964281962391008,
+ "wer": 2.304791830322074
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datatrota_mordtbrenner_1540-eval",
+ "label": "Workflow on data trota_mordtbrenner_1540",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/trota_mordtbrenner_1540.ocrd.zip",
+ "label": "GT workspace trota_mordtbrenner_1540"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/trota_mordtbrenner_1540_ocr.zip",
+ "label": "OCR workspace for trota_mordtbrenner_1540"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/trota_mordtbrenner_1540_evaluation.zip",
+ "label": "Evaluation workspace for trota_mordtbrenner_1540"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 6.38229,
+ "cpu_time": 7.811183,
+ "cer_mean": 0.043071000855431994,
+ "cer_median": 0.043071000855431994,
+ "cer_range": [
+ 0.014285714285714285,
+ 0.0718562874251497
+ ],
+ "cer_standard_deviation": 0.04070854266369089,
+ "wer": 0.10714905431199029,
+ "pages_per_minute": 18.802028738900926
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.0718562874251497,
+ "wer": 0.19469026548672566
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.014285714285714285,
+ "wer": 0.0196078431372549
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datajusti_abhandlung01_1758-eval",
+ "label": "Workflow on data justi_abhandlung01_1758",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/justi_abhandlung01_1758.ocrd.zip",
+ "label": "GT workspace justi_abhandlung01_1758"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/justi_abhandlung01_1758_ocr.zip",
+ "label": "OCR workspace for justi_abhandlung01_1758"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/justi_abhandlung01_1758_evaluation.zip",
+ "label": "Evaluation workspace for justi_abhandlung01_1758"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 27.452519000000002,
+ "cpu_time": 26.471052,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 8.742367139423527
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datahuebner_handbuch_1696-eval",
+ "label": "Workflow on data huebner_handbuch_1696",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/huebner_handbuch_1696.ocrd.zip",
+ "label": "GT workspace huebner_handbuch_1696"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/huebner_handbuch_1696_ocr.zip",
+ "label": "OCR workspace for huebner_handbuch_1696"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/huebner_handbuch_1696_evaluation.zip",
+ "label": "Evaluation workspace for huebner_handbuch_1696"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 3,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 48.470510000000004,
+ "cpu_time": 66.43220300000002,
+ "cer_mean": 0.20641950275814583,
+ "cer_median": 0.1099236641221374,
+ "cer_range": [
+ 0.09132841328413284,
+ 0.4180064308681672
+ ],
+ "cer_standard_deviation": 0.18347538513029096,
+ "wer": 0.3802990111723255,
+ "pages_per_minute": 3.713598227045682
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.4180064308681672,
+ "wer": 0.7073170731707317
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.1099236641221374,
+ "wer": 0.24778761061946902
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.09132841328413284,
+ "wer": 0.18579234972677597
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datakistler_kraeuter_1500-eval",
+ "label": "Workflow on data kistler_kraeuter_1500",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/kistler_kraeuter_1500.ocrd.zip",
+ "label": "GT workspace kistler_kraeuter_1500"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/kistler_kraeuter_1500_ocr.zip",
+ "label": "OCR workspace for kistler_kraeuter_1500"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/kistler_kraeuter_1500_evaluation.zip",
+ "label": "Evaluation workspace for kistler_kraeuter_1500"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 27.745072,
+ "cpu_time": 42.637415,
+ "cer_mean": 0.10034282802482036,
+ "cer_median": 0.10034282802482036,
+ "cer_range": [
+ 0.09958847736625515,
+ 0.10109717868338558
+ ],
+ "cer_standard_deviation": 0.001066812932128006,
+ "wer": 0.33029935275080907,
+ "pages_per_minute": 4.325092398390604
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0007",
+ "cer": 0.09958847736625515,
+ "wer": 0.36893203883495146
+ },
+ {
+ "page_id": "phys_0021",
+ "cer": 0.10109717868338558,
+ "wer": 0.2916666666666667
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataluz_blitz_1784-eval",
+ "label": "Workflow on data luz_blitz_1784",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/luz_blitz_1784.ocrd.zip",
+ "label": "GT workspace luz_blitz_1784"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luz_blitz_1784_ocr.zip",
+ "label": "OCR workspace for luz_blitz_1784"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luz_blitz_1784_evaluation.zip",
+ "label": "Evaluation workspace for luz_blitz_1784"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 17.701244,
+ "cpu_time": 20.535555,
+ "cer_mean": 0.030367084502750087,
+ "cer_median": 0.02381735856100435,
+ "cer_range": [
+ 0.014354066985645933,
+ 0.05947955390334572
+ ],
+ "cer_standard_deviation": 0.02030808283356641,
+ "wer": 0.08722583259487592,
+ "pages_per_minute": 13.558369118012271
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.05947955390334572,
+ "wer": 0.1794871794871795
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.014354066985645933,
+ "wer": 0.043478260869565216
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.02869097429766886,
+ "wer": 0.07924528301886792
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.01894374282433984,
+ "wer": 0.04669260700389105
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datahuebner_handbuch_1696-eval",
+ "label": "Workflow on data huebner_handbuch_1696",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/huebner_handbuch_1696.ocrd.zip",
+ "label": "GT workspace huebner_handbuch_1696"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/huebner_handbuch_1696_ocr.zip",
+ "label": "OCR workspace for huebner_handbuch_1696"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/huebner_handbuch_1696_evaluation.zip",
+ "label": "Evaluation workspace for huebner_handbuch_1696"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 3,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 28.173187000000002,
+ "cpu_time": 27.232793000000004,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 6.389053535192876
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datareichsanzeiger_title_pages-eval",
+ "label": "Workflow on data reichsanzeiger_title_pages",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_title_pages.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_title_pages"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_title_pages_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_title_pages"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_title_pages_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_title_pages"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-crop",
+ "params": {
+ "dpi": 0,
+ "padding": 4
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua"
+ ],
+ "publication_century": "1820-1939",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 5,
+ "layout": "reichsanzeiger-gt"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 1006.2491769999999,
+ "cpu_time": 965.354889,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 0.2981368898053768
+ },
+ "by_page": [
+ {
+ "page_id": "P_1881_115_0163",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1885_5_0054",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1887_134_0444",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1916_169_0087",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1918_267_0129",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datareichsanzeiger_many_ads-eval",
+ "label": "Workflow on data reichsanzeiger_many_ads",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_many_ads.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_many_ads"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_many_ads_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_many_ads"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_many_ads_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_many_ads"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua"
+ ],
+ "publication_century": "1820-1939",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 5,
+ "layout": "reichsanzeiger-gt"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 307.154189,
+ "cpu_time": 356.97703,
+ "cer_mean": 1.3340240550306781,
+ "cer_median": 0.7412443874278384,
+ "cer_range": [
+ 0.3960932753867003,
+ 3.7402255639097746
+ ],
+ "cer_standard_deviation": 1.378192503303974,
+ "wer": 1.8653811990174032,
+ "pages_per_minute": 0.9767081509671353
+ },
+ "by_page": [
+ {
+ "page_id": "P_1871_155_0279",
+ "cer": 0.3960932753867003,
+ "wer": 0.49564980967917344
+ },
+ {
+ "page_id": "P_1871_65_0045",
+ "cer": 0.7412443874278384,
+ "wer": 0.947814451382694
+ },
+ {
+ "page_id": "P_1873_1_0017",
+ "cer": 0.5852251348300515,
+ "wer": 0.6788418708240535
+ },
+ {
+ "page_id": "P_1881_1_0662",
+ "cer": 3.7402255639097746,
+ "wer": 5.734011627906977
+ },
+ {
+ "page_id": "P_1883_55_0044",
+ "cer": 1.2073319135990264,
+ "wer": 1.4705882352941178
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datablumenbach_anatomie_1805-eval",
+ "label": "Workflow on data blumenbach_anatomie_1805",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/blumenbach_anatomie_1805.ocrd.zip",
+ "label": "GT workspace blumenbach_anatomie_1805"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/blumenbach_anatomie_1805_ocr.zip",
+ "label": "OCR workspace for blumenbach_anatomie_1805"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/blumenbach_anatomie_1805_evaluation.zip",
+ "label": "Evaluation workspace for blumenbach_anatomie_1805"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
+ "model": "Fraktur_GT4HistOCR",
+ "dpi": 0,
+ "padding": 0,
+ "segmentation_level": "word",
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1800-1900",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 38.17563400000001,
+ "cpu_time": 59.410665,
+ "cer_mean": 0.08697690719872657,
+ "cer_median": 0.04421052631578947,
+ "cer_range": [
+ 0.03735325506937033,
+ 0.17936694021101993
+ ],
+ "cer_standard_deviation": 0.08008554296654442,
+ "wer": 0.2764482431149098,
+ "pages_per_minute": 4.715049395119409
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.04421052631578947,
+ "wer": 0.16666666666666666
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.03735325506937033,
+ "wer": 0.17037037037037037
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.17936694021101993,
+ "wer": 0.49230769230769234
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datareichsanzeiger_many_ads-eval",
+ "label": "Workflow on data reichsanzeiger_many_ads",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_many_ads.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_many_ads"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_many_ads_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_many_ads"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_many_ads_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_many_ads"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua"
+ ],
+ "publication_century": "1820-1939",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 5,
+ "layout": "reichsanzeiger-gt"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 1968.7780160000002,
+ "cpu_time": 1913.616593,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 0.15237878397764473
+ },
+ "by_page": [
+ {
+ "page_id": "P_1871_155_0279",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1871_65_0045",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1873_1_0017",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1881_1_0662",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1883_55_0044",
+ "cer": 1.0,
+ "wer": 1.0
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataweigel_gnothi02_1618-eval",
+ "label": "Workflow on data weigel_gnothi02_1618",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/weigel_gnothi02_1618.ocrd.zip",
+ "label": "GT workspace weigel_gnothi02_1618"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/weigel_gnothi02_1618_ocr.zip",
+ "label": "OCR workspace for weigel_gnothi02_1618"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/weigel_gnothi02_1618_evaluation.zip",
+ "label": "Evaluation workspace for weigel_gnothi02_1618"
+ },
+ "workflow_steps": [
+ {
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
+ }
+ }
+ ],
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 4,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 45.738777,
+ "cpu_time": 44.076152,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
+ "cer_range": [
+ 1.0,
+ 1.0
+ ],
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 5.247188834979125
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 1.0,
+ "wer": 1.0
}
]
}
},
{
- "eval_workflow_id": "wf-databernd_lebensbeschreibung_1738-eval",
- "label": "Workflow on data bernd_lebensbeschreibung_1738",
+ "eval_workflow_id": "wf-dataestor_rechtsgelehrsamkeit02_1758-eval",
+ "label": "Workflow on data estor_rechtsgelehrsamkeit02_1758",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
+ "label": "OCR Workflow selected_pages_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/bernd_lebensbeschreibung_1738.ocrd.zip",
- "label": "GT workspace bernd_lebensbeschreibung_1738"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/estor_rechtsgelehrsamkeit02_1758.ocrd.zip",
+ "label": "GT workspace estor_rechtsgelehrsamkeit02_1758"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bernd_lebensbeschreibung_1738_ocr.zip",
- "label": "OCR workspace for bernd_lebensbeschreibung_1738"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/estor_rechtsgelehrsamkeit02_1758_ocr.zip",
+ "label": "OCR workspace for estor_rechtsgelehrsamkeit02_1758"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bernd_lebensbeschreibung_1738_evaluation.zip",
- "label": "Evaluation workspace for bernd_lebensbeschreibung_1738"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/estor_rechtsgelehrsamkeit02_1758_evaluation.zip",
+ "label": "Evaluation workspace for estor_rechtsgelehrsamkeit02_1758"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
- "find_tables": true,
- "model": "Fraktur_GT4HistOCR",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
"dpi": 0,
- "padding": 0,
- "overwrite_segments": false,
- "overwrite_text": true,
- "shrink_polygons": false,
- "block_polygons": false,
- "find_staves": false,
- "sparse_text": false,
- "raw_lines": false,
- "char_whitelist": "",
- "char_blacklist": "",
- "char_unblacklist": "",
- "tesseract_parameters": {},
- "xpath_parameters": {},
- "xpath_model": {},
- "auto_model": false,
- "oem": "DEFAULT"
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-skimage-binarize",
+ "params": {
+ "method": "li",
+ "level-of-operation": "page",
+ "dpi": 0,
+ "window_size": 301,
+ "k": 0.34
+ }
+ },
+ {
+ "id": "ocrd-skimage-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "protect": 0.0,
+ "maxsize": 1.0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-segment",
+ "params": {
+ "dpi": 0,
+ "level-of-operation": "region",
+ "maxcolseps": 20,
+ "maxseps": 20,
+ "maximages": 10,
+ "csminheight": 4,
+ "hlminwidth": 10,
+ "gap_height": 0.01,
+ "gap_width": 1.5,
+ "overwrite_order": true,
+ "overwrite_separators": true,
+ "overwrite_regions": true,
+ "overwrite_lines": true,
+ "spread": 2.4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-calamari-recognize",
+ "params": {
+ "checkpoint_dir": "qurator-gt4histocr-1.0",
+ "voter": "confidence_voter_default_ctc",
+ "textequiv_level": "line",
+ "glyph_conf_cutoff": 0.001
}
}
],
- "workflow_model": "Fraktur_GT4HistOCR",
- "eval_tool": "ocrd-dinglehopper vNone",
+ "workflow_model": "qurator-gt4histocr-1.0",
+ "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.12.0)",
"document_metadata": {
"data_properties": {
"fonts": [
@@ -1825,47 +10947,52 @@
"publication_century": "1700-1800",
"publication_decade": "",
"publication_year": "18th century",
- "number_of_pages": 3,
+ "number_of_pages": 4,
"layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 8.577653,
- "cpu_time": 10.790615,
- "cer_mean": 0.1790109644215708,
- "cer_median": 0.05504587155963303,
+ "wall_time": 28.053300999999998,
+ "cpu_time": 26.997501,
+ "cer_mean": 1.0,
+ "cer_median": 1.0,
"cer_range": [
- 0.008409785932721712,
- 0.4735772357723577
+ 1.0,
+ 1.0
],
- "cer_standard_deviation": 0.2561653709691831,
- "wer": 0.2391410103864555,
- "pages_per_minute": 20.984761216150854
+ "cer_standard_deviation": 0.0,
+ "wer": 1.0,
+ "pages_per_minute": 8.555142940219406
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.4735772357723577,
- "wer": 0.5882352941176471
+ "cer": 1.0,
+ "wer": 1.0
},
{
"page_id": "phys_0002",
- "cer": 0.05504587155963303,
- "wer": 0.09821428571428571
+ "cer": 1.0,
+ "wer": 1.0
},
{
"page_id": "phys_0003",
- "cer": 0.008409785932721712,
- "wer": 0.030973451327433628
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 1.0,
+ "wer": 1.0
}
]
}
},
{
- "eval_workflow_id": "wf-dataweigel_gnothi02_1618-eval",
- "label": "Workflow on data weigel_gnothi02_1618",
+ "eval_workflow_id": "wf-databallenstedt_delatio_1777-eval",
+ "label": "Workflow on data ballenstedt_delatio_1777",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -1876,16 +11003,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/weigel_gnothi02_1618.ocrd.zip",
- "label": "GT workspace weigel_gnothi02_1618"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/ballenstedt_delatio_1777.ocrd.zip",
+ "label": "GT workspace ballenstedt_delatio_1777"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/weigel_gnothi02_1618_ocr.zip",
- "label": "OCR workspace for weigel_gnothi02_1618"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/ballenstedt_delatio_1777_ocr.zip",
+ "label": "OCR workspace for ballenstedt_delatio_1777"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/weigel_gnothi02_1618_evaluation.zip",
- "label": "Evaluation workspace for weigel_gnothi02_1618"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/ballenstedt_delatio_1777_evaluation.zip",
+ "label": "Evaluation workspace for ballenstedt_delatio_1777"
},
"workflow_steps": [
{
@@ -1923,55 +11050,50 @@
"Antiqua",
"Fraktur"
],
- "publication_century": "1600-1700",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "17th century",
- "number_of_pages": 4,
+ "publication_year": "18th century",
+ "number_of_pages": 3,
"layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 27.856097,
- "cpu_time": 31.285033,
- "cer_mean": 0.11399763421865425,
- "cer_median": 0.09987627660053089,
+ "wall_time": 18.330509,
+ "cpu_time": 20.979165,
+ "cer_mean": 0.06438361522903834,
+ "cer_median": 0.03969957081545064,
"cer_range": [
- 0.07531106745252128,
- 0.18092691622103388
+ 0.021764032073310423,
+ 0.13168724279835392
],
- "cer_standard_deviation": 0.047514032403792344,
- "wer": 0.19377061794956923,
- "pages_per_minute": 8.615708079994121
+ "cer_standard_deviation": 0.058972490200809365,
+ "wer": 0.16906902212925057,
+ "pages_per_minute": 9.819694586767886
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.18092691622103388,
- "wer": 0.3282051282051282
- },
- {
- "page_id": "phys_0002",
- "cer": 0.07531106745252128,
- "wer": 0.15019762845849802
+ "page_id": "phys_00003",
+ "cer": 0.021764032073310423,
+ "wer": 0.12236286919831224
},
{
- "page_id": "phys_0003",
- "cer": 0.11397849462365592,
- "wer": 0.15934065934065933
+ "page_id": "phys_00005",
+ "cer": 0.13168724279835392,
+ "wer": 0.2603305785123967
},
{
- "page_id": "phys_0004",
- "cer": 0.08577405857740586,
- "wer": 0.13733905579399142
+ "page_id": "phys_00010",
+ "cer": 0.03969957081545064,
+ "wer": 0.1245136186770428
}
]
}
},
{
- "eval_workflow_id": "wf-datalessing_menschengeschlecht_1780-eval",
- "label": "Workflow on data lessing_menschengeschlecht_1780",
+ "eval_workflow_id": "wf-datareichsanzeiger_title_pages-eval",
+ "label": "Workflow on data reichsanzeiger_title_pages",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -1982,16 +11104,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/lessing_menschengeschlecht_1780.ocrd.zip",
- "label": "GT workspace lessing_menschengeschlecht_1780"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_title_pages.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_title_pages"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/lessing_menschengeschlecht_1780_ocr.zip",
- "label": "OCR workspace for lessing_menschengeschlecht_1780"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_title_pages_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_title_pages"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/lessing_menschengeschlecht_1780_evaluation.zip",
- "label": "Evaluation workspace for lessing_menschengeschlecht_1780"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_title_pages_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_title_pages"
},
"workflow_steps": [
{
@@ -2026,78 +11148,177 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua",
- "Fraktur"
+ "Antiqua"
],
- "publication_century": "1700-1800",
+ "publication_century": "1820-1939",
"publication_decade": "",
- "publication_year": "18th century",
- "number_of_pages": 1,
- "layout": "simple"
+ "publication_year": "19th century",
+ "number_of_pages": 5,
+ "layout": "reichsanzeiger-gt"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 2.026755,
- "cpu_time": 2.56115,
- "cer_mean": 0.02493765586034913,
- "cer_median": 0.02493765586034913,
+ "wall_time": 365.603086,
+ "cpu_time": 453.722946,
+ "cer_mean": 0.5074594248191785,
+ "cer_median": 0.3389393598117977,
"cer_range": [
- 0.02493765586034913,
- 0.02493765586034913
+ 0.09154875911296263,
+ 1.0
],
- "cer_standard_deviation": null,
- "wer": 0.09836065573770492,
- "pages_per_minute": 29.60397285315689
+ "cer_standard_deviation": 0.36783069273412317,
+ "wer": 0.630009265947987,
+ "pages_per_minute": 0.8205620014925147
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.02493765586034913,
- "wer": 0.09836065573770492
+ "page_id": "P_1881_115_0163",
+ "cer": 0.3377672361293631,
+ "wer": 0.4791666666666667
+ },
+ {
+ "page_id": "P_1885_5_0054",
+ "cer": 0.09154875911296263,
+ "wer": 0.20966135458167331
+ },
+ {
+ "page_id": "P_1887_134_0444",
+ "cer": 1.0,
+ "wer": 1.0
+ },
+ {
+ "page_id": "P_1916_169_0087",
+ "cer": 0.3389393598117977,
+ "wer": 0.5072414380644062
+ },
+ {
+ "page_id": "P_1918_267_0129",
+ "cer": 0.769041769041769,
+ "wer": 0.953976870427189
}
]
}
},
{
- "eval_workflow_id": "wf-dataalberti_pictura_1540-eval",
- "label": "Workflow on data alberti_pictura_1540",
+ "eval_workflow_id": "wf-dataeuler_rechenkunst01_1738-eval",
+ "label": "Workflow on data euler_rechenkunst01_1738",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/alberti_pictura_1540.ocrd.zip",
- "label": "GT workspace alberti_pictura_1540"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/euler_rechenkunst01_1738.ocrd.zip",
+ "label": "GT workspace euler_rechenkunst01_1738"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/alberti_pictura_1540_ocr.zip",
- "label": "OCR workspace for alberti_pictura_1540"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/euler_rechenkunst01_1738_ocr.zip",
+ "label": "OCR workspace for euler_rechenkunst01_1738"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/alberti_pictura_1540_evaluation.zip",
- "label": "Evaluation workspace for alberti_pictura_1540"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/euler_rechenkunst01_1738_evaluation.zip",
+ "label": "Evaluation workspace for euler_rechenkunst01_1738"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
+ "params": {
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
"find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "overwrite_segments": false,
+ "segmentation_level": "word",
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -2120,50 +11341,65 @@
"Antiqua",
"Fraktur"
],
- "publication_century": "1500-1600",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "16th century",
- "number_of_pages": 3,
+ "publication_year": "18th century",
+ "number_of_pages": 6,
"layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 7.85773,
- "cpu_time": 9.961092,
- "cer_mean": 0.10240852523716282,
- "cer_median": 0.10536980749746708,
+ "wall_time": 69.019748,
+ "cpu_time": 108.522222,
+ "cer_mean": 0.1500034071140843,
+ "cer_median": 0.12202020202020203,
"cer_range": [
- 0.07124352331606218,
- 0.1306122448979592
+ 0.03902862098872507,
+ 0.39473684210526316
],
- "cer_standard_deviation": 0.02979493530847308,
- "wer": 0.23466068901129858,
- "pages_per_minute": 22.907379103125205
+ "cer_standard_deviation": 0.12934283380557174,
+ "wer": 0.27891577144281926,
+ "pages_per_minute": 5.215898499078843
},
"by_page": [
{
- "page_id": "phys_0007",
- "cer": 0.07124352331606218,
- "wer": 0.2231404958677686
+ "page_id": "phys_0001",
+ "cer": 0.03902862098872507,
+ "wer": 0.0872093023255814
},
{
- "page_id": "phys_0008",
- "cer": 0.10536980749746708,
- "wer": 0.2484472049689441
+ "page_id": "phys_0002",
+ "cer": 0.39473684210526316,
+ "wer": 0.6402439024390244
},
{
- "page_id": "phys_0009",
- "cer": 0.1306122448979592,
- "wer": 0.2323943661971831
+ "page_id": "phys_0003",
+ "cer": 0.16756341275941583,
+ "wer": 0.3592233009708738
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.13737373737373737,
+ "wer": 0.2
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 0.10666666666666667,
+ "wer": 0.21893491124260356
+ },
+ {
+ "page_id": "phys_0006",
+ "cer": 0.05465116279069768,
+ "wer": 0.1678832116788321
}
]
}
},
{
- "eval_workflow_id": "wf-dataheyden_paedono_1548-eval",
- "label": "Workflow on data heyden_paedono_1548",
+ "eval_workflow_id": "wf-dataalberti_pictura_1540-eval",
+ "label": "Workflow on data alberti_pictura_1540",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -2174,16 +11410,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/heyden_paedono_1548.ocrd.zip",
- "label": "GT workspace heyden_paedono_1548"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/alberti_pictura_1540.ocrd.zip",
+ "label": "GT workspace alberti_pictura_1540"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/heyden_paedono_1548_ocr.zip",
- "label": "OCR workspace for heyden_paedono_1548"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/alberti_pictura_1540_ocr.zip",
+ "label": "OCR workspace for alberti_pictura_1540"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/heyden_paedono_1548_evaluation.zip",
- "label": "Evaluation workspace for heyden_paedono_1548"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/alberti_pictura_1540_evaluation.zip",
+ "label": "Evaluation workspace for alberti_pictura_1540"
},
"workflow_steps": [
{
@@ -2225,46 +11461,46 @@
"publication_decade": "",
"publication_year": "16th century",
"number_of_pages": 3,
- "layout": "simple"
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 8.303854,
- "cpu_time": 10.292532,
- "cer_mean": 0.07452119312897007,
- "cer_median": 0.0629800307219662,
+ "wall_time": 7.080852,
+ "cpu_time": 9.399835,
+ "cer_mean": 0.10240852523716282,
+ "cer_median": 0.10536980749746708,
"cer_range": [
- 0.037037037037037035,
- 0.12354651162790697
+ 0.07124352331606218,
+ 0.1306122448979592
],
- "cer_standard_deviation": 0.044394494261965886,
- "wer": 0.22683890077340793,
- "pages_per_minute": 21.676681695029803
+ "cer_standard_deviation": 0.02979493530847308,
+ "wer": 0.23466068901129858,
+ "pages_per_minute": 25.420669716017226
},
"by_page": [
{
"page_id": "phys_0007",
- "cer": 0.0629800307219662,
- "wer": 0.21359223300970873
+ "cer": 0.07124352331606218,
+ "wer": 0.2231404958677686
},
{
- "page_id": "phys_0013",
- "cer": 0.12354651162790697,
- "wer": 0.33980582524271846
+ "page_id": "phys_0008",
+ "cer": 0.10536980749746708,
+ "wer": 0.2484472049689441
},
{
- "page_id": "phys_0014",
- "cer": 0.037037037037037035,
- "wer": 0.1271186440677966
+ "page_id": "phys_0009",
+ "cer": 0.1306122448979592,
+ "wer": 0.2323943661971831
}
]
}
},
{
- "eval_workflow_id": "wf-dataluz_blitz_1784-eval",
- "label": "Workflow on data luz_blitz_1784",
+ "eval_workflow_id": "wf-datann_besuch_1780-eval",
+ "label": "Workflow on data nn_besuch_1780",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -2275,16 +11511,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/luz_blitz_1784.ocrd.zip",
- "label": "GT workspace luz_blitz_1784"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/nn_besuch_1780.ocrd.zip",
+ "label": "GT workspace nn_besuch_1780"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luz_blitz_1784_ocr.zip",
- "label": "OCR workspace for luz_blitz_1784"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/nn_besuch_1780_ocr.zip",
+ "label": "OCR workspace for nn_besuch_1780"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luz_blitz_1784_evaluation.zip",
- "label": "Evaluation workspace for luz_blitz_1784"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/nn_besuch_1780_evaluation.zip",
+ "label": "Evaluation workspace for nn_besuch_1780"
},
"workflow_steps": [
{
@@ -2332,82 +11568,116 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 19.577649,
- "cpu_time": 22.415525,
- "cer_mean": 0.030367084502750087,
- "cer_median": 0.02381735856100435,
+ "wall_time": 12.750288,
+ "cpu_time": 15.089726,
+ "cer_mean": 0.06315407734401027,
+ "cer_median": 0.026726016076928857,
"cer_range": [
- 0.014354066985645933,
- 0.05947955390334572
+ 0.01878453038674033,
+ 0.18037974683544303
],
- "cer_standard_deviation": 0.02030808283356641,
- "wer": 0.08722583259487592,
- "pages_per_minute": 12.258877457655922
+ "cer_standard_deviation": 0.07825196427362012,
+ "wer": 0.11662846322503488,
+ "pages_per_minute": 18.82310423105737
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.05947955390334572,
- "wer": 0.1794871794871795
+ "page_id": "phys_00001",
+ "cer": 0.18037974683544303,
+ "wer": 0.1896551724137931
},
{
- "page_id": "phys_0002",
- "cer": 0.014354066985645933,
- "wer": 0.043478260869565216
+ "page_id": "phys_00002",
+ "cer": 0.01878453038674033,
+ "wer": 0.08670520231213873
},
{
- "page_id": "phys_0003",
- "cer": 0.02869097429766886,
- "wer": 0.07924528301886792
+ "page_id": "phys_00003",
+ "cer": 0.02505446623093682,
+ "wer": 0.08670520231213873
},
{
- "page_id": "phys_0004",
- "cer": 0.01894374282433984,
- "wer": 0.04669260700389105
+ "page_id": "phys_00004",
+ "cer": 0.028397565922920892,
+ "wer": 0.10344827586206896
}
]
}
},
{
- "eval_workflow_id": "wf-dataestor_rechtsgelehrsamkeit02_1758-eval",
- "label": "Workflow on data estor_rechtsgelehrsamkeit02_1758",
+ "eval_workflow_id": "wf-datareichsanzeiger_tables-eval",
+ "label": "Workflow on data reichsanzeiger_tables",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/estor_rechtsgelehrsamkeit02_1758.ocrd.zip",
- "label": "GT workspace estor_rechtsgelehrsamkeit02_1758"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_tables.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_tables"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/estor_rechtsgelehrsamkeit02_1758_ocr.zip",
- "label": "OCR workspace for estor_rechtsgelehrsamkeit02_1758"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_tables_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_tables"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/estor_rechtsgelehrsamkeit02_1758_evaluation.zip",
- "label": "Evaluation workspace for estor_rechtsgelehrsamkeit02_1758"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_tables_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_tables"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
- "find_tables": true,
- "model": "Fraktur_GT4HistOCR",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
"dpi": 0,
- "padding": 0,
- "overwrite_segments": false,
- "overwrite_text": true,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-crop",
+ "params": {
+ "dpi": 0,
+ "padding": 4
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
"raw_lines": false,
"char_whitelist": "",
"char_blacklist": "",
@@ -2416,102 +11686,31 @@
"xpath_parameters": {},
"xpath_model": {},
"auto_model": false,
- "oem": "DEFAULT"
- }
- }
- ],
- "workflow_model": "Fraktur_GT4HistOCR",
- "eval_tool": "ocrd-dinglehopper vNone",
- "document_metadata": {
- "data_properties": {
- "fonts": [
- "Antiqua",
- "Fraktur"
- ],
- "publication_century": "1700-1800",
- "publication_decade": "",
- "publication_year": "18th century",
- "number_of_pages": 4,
- "layout": "complex"
- }
- }
- },
- "evaluation_results": {
- "document_wide": {
- "wall_time": 19.327455,
- "cpu_time": 22.561449,
- "cer_mean": 0.11589408928020027,
- "cer_median": 0.10084584323499293,
- "cer_range": [
- 0.05331088664421998,
- 0.2085737840065952
- ],
- "cer_standard_deviation": 0.07259145757108061,
- "wer": 0.20102650242627845,
- "pages_per_minute": 12.417568686617043
- },
- "by_page": [
- {
- "page_id": "phys_0001",
- "cer": 0.06310013717421124,
- "wer": 0.11403508771929824
- },
- {
- "page_id": "phys_0002",
- "cer": 0.13859154929577464,
- "wer": 0.22340425531914893
+ "oem": "DEFAULT"
+ }
},
{
- "page_id": "phys_0003",
- "cer": 0.05331088664421998,
- "wer": 0.13333333333333333
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
},
- {
- "page_id": "phys_0004",
- "cer": 0.2085737840065952,
- "wer": 0.3333333333333333
- }
- ]
- }
- },
- {
- "eval_workflow_id": "wf-datatrota_mordtbrenner_1540-eval",
- "label": "Workflow on data trota_mordtbrenner_1540",
- "metadata": {
- "ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
- },
- "eval_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
- "label": "Evaluation Workflow dinglehopper_eval"
- },
- "gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/trota_mordtbrenner_1540.ocrd.zip",
- "label": "GT workspace trota_mordtbrenner_1540"
- },
- "ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/trota_mordtbrenner_1540_ocr.zip",
- "label": "OCR workspace for trota_mordtbrenner_1540"
- },
- "eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/trota_mordtbrenner_1540_evaluation.zip",
- "label": "Evaluation workspace for trota_mordtbrenner_1540"
- },
- "workflow_steps": [
{
"id": "ocrd-tesserocr-recognize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
- "find_tables": true,
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "overwrite_segments": false,
+ "segmentation_level": "word",
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -2531,41 +11730,55 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua",
- "Fraktur"
+ "Antiqua"
],
- "publication_century": "1500-1600",
+ "publication_century": "1820-1939",
"publication_decade": "",
- "publication_year": "16th century",
- "number_of_pages": 2,
- "layout": "simple"
+ "publication_year": "19th century",
+ "number_of_pages": 5,
+ "layout": "reichsanzeiger-gt"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 7.192959,
- "cpu_time": 8.533417,
- "cer_mean": 0.043071000855431994,
- "cer_median": 0.043071000855431994,
+ "wall_time": 4081.6259200000004,
+ "cpu_time": 4975.624208,
+ "cer_mean": 1.3864721075704818,
+ "cer_median": 0.9776101627630012,
"cer_range": [
- 0.014285714285714285,
- 0.0718562874251497
+ 0.9091116519622756,
+ 2.105984787830264
],
- "cer_standard_deviation": 0.04070854266369089,
- "wer": 0.10714905431199029,
- "pages_per_minute": 16.682981232063188
+ "cer_standard_deviation": 0.6085337634117919,
+ "wer": 1.6178586644645931,
+ "pages_per_minute": 0.0735001212458
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.0718562874251497,
- "wer": 0.19469026548672566
+ "page_id": "P_1877_7_0059",
+ "cer": 0.9432257390577666,
+ "wer": 0.9872849663425579
},
{
- "page_id": "phys_0002",
- "cer": 0.014285714285714285,
- "wer": 0.0196078431372549
+ "page_id": "P_1883_55_0044",
+ "cer": 0.9091116519622756,
+ "wer": 0.9841819080573406
+ },
+ {
+ "page_id": "P_1929_250_0019",
+ "cer": 0.9776101627630012,
+ "wer": 0.9989969909729187
+ },
+ {
+ "page_id": "P_1932_300_0488",
+ "cer": 2.105984787830264,
+ "wer": 2.814037626628075
+ },
+ {
+ "page_id": "P_1936_123_0292",
+ "cer": 1.9964281962391008,
+ "wer": 2.304791830322074
}
]
}
@@ -2575,8 +11788,8 @@
"label": "Workflow on data justi_abhandlung01_1758",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
@@ -2596,18 +11809,98 @@
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
"find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "overwrite_segments": false,
+ "segmentation_level": "word",
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -2640,80 +11933,160 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 13.437096,
- "cpu_time": 16.64254,
- "cer_mean": 0.12942563348778183,
- "cer_median": 0.13278777025964683,
+ "wall_time": 51.503143,
+ "cpu_time": 81.95352600000001,
+ "cer_mean": 0.06272817755704542,
+ "cer_median": 0.04686327354908531,
"cer_range": [
- 0.09893550407013149,
+ 0.0039946737683089215,
0.15319148936170213
],
- "cer_standard_deviation": 0.022501323128153037,
- "wer": 0.19483806092245076,
- "pages_per_minute": 17.861002109384348
+ "cer_standard_deviation": 0.07133216362479361,
+ "wer": 0.0881207064150021,
+ "pages_per_minute": 4.6599097845348965
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.13375130616509928,
- "wer": 0.2
+ "cer": 0.0073145245559038665,
+ "wer": 0.02857142857142857
},
{
"page_id": "phys_0002",
"cer": 0.15319148936170213,
- "wer": 0.22916666666666666
+ "wer": 0.20833333333333334
},
{
"page_id": "phys_0003",
- "cer": 0.09893550407013149,
- "wer": 0.12992125984251968
+ "cer": 0.08641202254226675,
+ "wer": 0.10236220472440945
},
{
"page_id": "phys_0004",
- "cer": 0.1318242343541944,
- "wer": 0.22026431718061673
+ "cer": 0.0039946737683089215,
+ "wer": 0.013215859030837005
}
]
}
},
{
- "eval_workflow_id": "wf-datareichsanzeiger_random-eval",
- "label": "Workflow on data reichsanzeiger_random",
+ "eval_workflow_id": "wf-dataarnimb_goethe03_1835-eval",
+ "label": "Workflow on data arnimb_goethe03_1835",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
- "label": "OCR Workflow minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
+ "label": "OCR Workflow slower_processors_ocr"
},
"eval_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_random.ocrd.zip",
- "label": "GT workspace reichsanzeiger_random"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/arnimb_goethe03_1835.ocrd.zip",
+ "label": "GT workspace arnimb_goethe03_1835"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_random_ocr.zip",
- "label": "OCR workspace for reichsanzeiger_random"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/arnimb_goethe03_1835_ocr.zip",
+ "label": "OCR workspace for arnimb_goethe03_1835"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_random_evaluation.zip",
- "label": "Evaluation workspace for reichsanzeiger_random"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/arnimb_goethe03_1835_evaluation.zip",
+ "label": "Evaluation workspace for arnimb_goethe03_1835"
},
"workflow_steps": [
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-binarize",
"params": {
- "segmentation_level": "region",
- "textequiv_level": "word",
+ "method": "ocropy",
+ "threshold": 0.5,
+ "grayscale": false,
+ "maxskew": 0.0,
+ "noise_maxsize": 0,
+ "dpi": 0,
+ "level-of-operation": "page"
+ }
+ },
+ {
+ "id": "ocrd-anybaseocr-crop",
+ "params": {
+ "dpi": 0,
+ "rulerRatioMax": 50.0,
+ "rulerRatioMin": 3.0,
+ "rulerAreaMax": 0.3,
+ "rulerAreaMin": 0.01,
+ "rulerWidthMax": 0.95,
+ "columnAreaMin": 0.05,
+ "columnSepWidthMax": 0.04,
+ "marginTop": 0.25,
+ "marginBottom": 0.75,
+ "marginLeft": 0.3,
+ "marginRight": 0.7,
+ "padding": 10
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-denoise",
+ "params": {
+ "level-of-operation": "page",
+ "noise_maxsize": 3.0,
+ "dpi": 0
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-deskew",
+ "params": {
+ "operation_level": "page",
+ "dpi": 0,
+ "min_orientation_confidence": 1.5
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-segment",
+ "params": {
+ "dpi": 0,
+ "padding": 4,
+ "shrink_polygons": false,
+ "block_polygons": false,
"find_tables": true,
+ "find_staves": false,
+ "sparse_text": false,
+ "overwrite_segments": true,
+ "segmentation_level": "region",
+ "textequiv_level": "none",
+ "overwrite_text": true,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
+ }
+ },
+ {
+ "id": "ocrd-cis-ocropy-dewarp",
+ "params": {
+ "dpi": 0,
+ "range": 4.0,
+ "smoothness": 1.0,
+ "max_neighbour": 0.05
+ }
+ },
+ {
+ "id": "ocrd-tesserocr-recognize",
+ "params": {
+ "textequiv_level": "glyph",
+ "overwrite_segments": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "overwrite_segments": false,
+ "segmentation_level": "word",
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
+ "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -2733,60 +12106,35 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua"
+ "Fraktur"
],
- "publication_century": "1820-1939",
+ "publication_century": "1800-1900",
"publication_decade": "",
"publication_year": "19th century",
- "number_of_pages": 6,
- "layout": "reichsanzeiger-gt"
+ "number_of_pages": 1,
+ "layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 636.322401,
- "cpu_time": 852.529652,
- "cer_mean": 1.1094634128044334,
- "cer_median": 0.8993673021608106,
+ "wall_time": 11.160878999999998,
+ "cpu_time": 17.708654999999997,
+ "cer_mean": 0.0056657223796034,
+ "cer_median": 0.0056657223796034,
"cer_range": [
- 0.30019453260980855,
- 1.9778823058446757
+ 0.0056657223796034,
+ 0.0056657223796034
],
- "cer_standard_deviation": 0.645494993792005,
- "wer": 1.3304151477151094,
- "pages_per_minute": 0.565750945486516
+ "cer_standard_deviation": null,
+ "wer": 0.031746031746031744,
+ "pages_per_minute": 5.3759206600125315
},
"by_page": [
{
- "page_id": "P_1879_45_0344",
- "cer": 0.9843462873477166,
- "wer": 1.1759884281581485
- },
- {
- "page_id": "P_1885_5_0055",
- "cer": 0.30019453260980855,
- "wer": 0.40127817019845274
- },
- {
- "page_id": "P_1889_1_0018",
- "cer": 0.7868184930272782,
- "wer": 0.9704142011834319
- },
- {
- "page_id": "P_1891_33_0452",
- "cer": 0.8143883169739047,
- "wer": 1.112810707456979
- },
- {
- "page_id": "P_1932_300_0488",
- "cer": 1.9778823058446757,
- "wer": 2.349493487698987
- },
- {
- "page_id": "P_1936_123_0292",
- "cer": 1.7931505410232167,
- "wer": 1.9725058915946583
+ "page_id": "phys_0001",
+ "cer": 0.0056657223796034,
+ "wer": 0.031746031746031744
}
]
}
diff --git a/src/benchmark_extraction.py b/src/benchmark_extraction.py
index 39b4b2144d236dec217160b415ca978ff35661d0..1c030c87f769d43f8fb29b2a745cfce5fc8fe7f4 100644
--- a/src/benchmark_extraction.py
+++ b/src/benchmark_extraction.py
@@ -7,7 +7,6 @@ import xml.etree.ElementTree as ET
from os import listdir, scandir
from statistics import stdev, median
from typing import Any, Dict, List, Union
-
from .constants import METS, RESULTS, QUIVER_MAIN, OCRD
@@ -198,8 +197,11 @@ def get_nextflow_time(workspace_path: str, time_type: str) -> float:
for log in logs:
with open(highest_workspace_dir + '/' + log, 'r', encoding='utf-8') as l:
log_file = l.read()
- no_sec_s = re.search(rf'([0-9]+?\.[0-9]+?)s \({time_type}\)', log_file).group(1)
- time_per_workflow_step.append(float(no_sec_s))
+ try:
+ no_sec_s = re.search(rf'([0-9]+?\.[0-9]+?)s \({time_type}\)', log_file).group(1)
+ time_per_workflow_step.append(float(no_sec_s))
+ except AttributeError:
+ print(f'No wall time found in {highest_workspace_dir}/{log}. Skipping.')
return sum(time_per_workflow_step)
@@ -207,7 +209,11 @@ def get_pages_per_minute(workspace_path: str) -> float:
duration = get_nextflow_time(workspace_path, 'wall')
no_pages = get_no_of_pages(workspace_path)
- return no_pages / (duration / 60)
+ try:
+ return no_pages / (duration / 60)
+ except ZeroDivisionError:
+ print('ERROR: Division by zero.')
+ return None
def get_mean_cer(workspace_path: str, gt_type: str) -> float:
diff --git a/workflows/execute_workflows.sh b/workflows/execute_workflows.sh
index 3ce0d9bf5a8d5f87838c03564eb6c45ceefa7803..5a90275622ce44f0a5c38c009840a5eba912bec0 100755
--- a/workflows/execute_workflows.sh
+++ b/workflows/execute_workflows.sh
@@ -69,28 +69,30 @@ create_wf_specific_workspaces() {
execute_wfs_and_extract_benchmarks() {
# for all data sets…
- for WS_DIR in "$WORKSPACE_DIR"/*
+ for WS_DIR in "$WORKSPACE_DIR"/*/
do
- DATA_DIR="$WS_DIR"/data
- DIR_NAME=$(basename "$WS_DIR")
- INNER_DIR=$(ls "$DATA_DIR"/)
-
- if ! grep -q "OCR-D-OCR" "$WS_DIR/data/$INNER_DIR/mets.xml" ; then
- echo "Switching to $WS_DIR."
-
- run "$DATA_DIR"/*/*ocr.txt.nf "$DIR_NAME" "$WS_DIR"
- run "$DATA_DIR"/*/*eval.txt.nf "$DIR_NAME" "$WS_DIR"
-
- # create a result JSON according to the specs
- echo "Get Benchmark JSON …"
- WORKFLOW=$(basename -s .txt.nf "$DATA_DIR"/*/*ocr.txt.nf)
- quiver benchmarks-extraction "$DATA_DIR"/* "$WORKFLOW"
- echo "Done."
-
- # move data to results dir
- mv "$DATA_DIR"/*/*result.json "$RESULTS_DIR"
- else
- echo "$WS_DIR has already been processed."
+ if [ "$WS_DIR" != "/app/workflows/workspaces/work/" ]; then
+ DATA_DIR="$WS_DIR"/data
+ DIR_NAME=$(basename "$WS_DIR")
+ INNER_DIR=$(ls "$DATA_DIR"/)
+
+ if ! grep -q "OCR-D-OCR" "$WS_DIR/data/$INNER_DIR/mets.xml"; then
+ echo "Switching to $WS_DIR."
+
+ run "$DATA_DIR"/*/*ocr.txt.nf "$DIR_NAME"
+ run "$DATA_DIR"/*/*eval.txt.nf "$DIR_NAME"
+
+ # create a result JSON according to the specs
+ echo "Get Benchmark JSON …"
+ WORKFLOW=$(basename -s .txt.nf "$DATA_DIR"/*/*ocr.txt.nf)
+ quiver benchmarks-extraction "$WS_DIR"/data/* "$WORKFLOW"
+ echo "Done."
+
+ # move data to results dir
+ mv "$DATA_DIR"/*/*result.json "$RESULTS_DIR"
+ else
+ echo "$WS_DIR has already been processed."
+ fi
fi
done
cd "$ROOT" || exit
@@ -120,20 +122,22 @@ rename_and_move_nextflow_result() {
run() {
# $1: $WORKFLOW
# $2: $DIR_NAME
- # $3: $WS_DIR
nextflow run "$1" -with-weblog http://127.0.0.1:8000/nextflow/ --mets_path "/app/workflows/workspaces/$2/data/*/mets.xml"
rename_and_move_nextflow_result "$1" "$2"
- save_workspaces "$3"/data "$2" "$1"
+ save_workspaces "$1" "$2"
}
save_workspaces() {
# $1: $WORKFLOW
# $2: $DIR_NAME
- # $3: $WS_DIR
- echo "Zipping workspace $3"
- ocrd -l ERROR zip bag -d "$DIR_NAME"/data/* -i "$DIR_NAME"/data/* "$DIR_NAME"
- WORKFLOW_NAME=$(basename -s .txt.nf "$1")
- mv "$WORKSPACE_DIR"/"$2".zip "$RESULTS_DIR"/"$2"_"$WORKFLOW_NAME".zip
+ echo "Zipping workspace $2"
+ DATA_DIR="$2/data/"
+ if basename -s .txt.nf "$1" | grep "eval"; then
+ WORKFLOW_NAME=$(basename -s .txt.nf "$1")
+ ocrd -l ERROR zip bag -d "$DATA_DIR"/* -i "$DATA_DIR"/* "$RESULTS_DIR"/"$2"_"$WORKFLOW_NAME".zip
+ else
+ ocrd -l ERROR zip bag -d "$DATA_DIR"/* -i "$DATA_DIR"/* "$RESULTS_DIR"/"$2".zip
+ fi
}
summarize_to_data_json() {
diff --git a/workflows/ocrd_workflows/selected_pages_ocr.txt b/workflows/ocrd_workflows/selected_pages_ocr.txt
index a6c3c7901b4858b403bf6f95fa328d1ae6c46d5b..864a00e85b3635ca22f0ac27389b1e3fe5d03443 100644
--- a/workflows/ocrd_workflows/selected_pages_ocr.txt
+++ b/workflows/ocrd_workflows/selected_pages_ocr.txt
@@ -1,6 +1,6 @@
ocrd process \
"cis-ocropy-binarize -I OCR-D-IMG -O OCR-D-BIN" \
- "anybaseocr-crop -I OCR-D-BIN -O OCR-D-CROP" \
+ "tesserocr-crop -I OCR-D-BIN -O OCR-D-CROP" \
"skimage-binarize -I OCR-D-CROP -O OCR-D-BIN2 -P method li" \
"skimage-denoise -I OCR-D-BIN2 -O OCR-D-BIN-DENOISE -P level-of-operation page" \
"tesserocr-deskew -I OCR-D-BIN-DENOISE -O OCR-D-BIN-DENOISE-DESKEW -P operation_level page" \
diff --git a/workflows/ocrd_workflows/slower_processors_ocr.txt b/workflows/ocrd_workflows/slower_processors_ocr.txt
index 299ef9b03ee20ce93a02bba00a7c7f2b7b36946c..5cabe322e859ef7984f0296d3b949de44df565df 100644
--- a/workflows/ocrd_workflows/slower_processors_ocr.txt
+++ b/workflows/ocrd_workflows/slower_processors_ocr.txt
@@ -1,6 +1,6 @@
ocrd process \
"cis-ocropy-binarize -I OCR-D-IMG -O OCR-D-BIN" \
- "anybaseocr-crop -I OCR-D-BIN -O OCR-D-CROP" \
+ "tesserocr-crop -I OCR-D-BIN -O OCR-D-CROP" \
"cis-ocropy-denoise -I OCR-D-CROP -O OCR-D-BIN-DENOISE -P level-of-operation page" \
"tesserocr-deskew -I OCR-D-BIN-DENOISE -O OCR-D-BIN-DENOISE-DESKEW -P operation_level page" \
"tesserocr-segment -I OCR-D-BIN-DENOISE-DESKEW -O OCR-D-SEG" \