diff --git a/.gitignore b/.gitignore
index 4da1598e9800433f085c4fabf79c2be8c5b16df3..77ec93a82bff41fe605047ced6a7df55ffe8a490 100644
--- a/.gitignore
+++ b/.gitignore
@@ -8,4 +8,5 @@ workflows/nf-results/*
workflows/results
workflows/ocrd-workflows/*.nf
models
-.idea
\ No newline at end of file
+.idea
+gt/*
diff --git a/Dockerfile b/Dockerfile
index 02321b76f01e828150eeb5dfc9503c54592d221f..ae906dfc8312def615efdadc8008fa69256b3991 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -5,6 +5,7 @@ WORKDIR /app
COPY requirements.txt requirements.txt
RUN apt install git
+RUN apt install -y jq
RUN apt-get update
RUN apt-get install -y --fix-missing openjdk-11-jre
@@ -14,7 +15,6 @@ COPY README.md README.md
RUN git init
RUN git submodule add https://github.com/MehmedGIT/OtoN_Converter submodules/oton
-RUN git submodule add https://github.com/OCR-D/quiver-data submodules/quiver-data
RUN git submodule update --init
RUN cd submodules/oton && \
@@ -22,10 +22,13 @@ RUN cd submodules/oton && \
sed -i "s \$projectDir/ocrd-workspace/ $WORKSPACE_DIR/CURRENT/ g" oton/config.toml && \
pip install .
+COPY prepare.sh prepare.sh
+COPY default_data_sources.txt default_data_sources.txt
+
RUN pip3 install -r requirements.txt
RUN pip3 install .
RUN nextflow
COPY workflows workflows
-CMD [ "bash", "workflows/execute_workflows.sh" ]
\ No newline at end of file
+ENTRYPOINT [ "bash" ]
\ No newline at end of file
diff --git a/README.md b/README.md
index 5f5f3f977b9a1f0a7866e33564b0665d67f98d27..98e95f0f07e5af4aea4ff5de44c4c8dd3653981c 100644
--- a/README.md
+++ b/README.md
@@ -11,7 +11,7 @@ QuiVer Benchmarks is based on `ocrd/all:maximum` and has all OCR-D processors at
- Docker >= 23.0.0
- [Docker Compose plugin](https://docs.docker.com/compose/install/linux/#install-using-the-repository)
-To speed up QuiVer Benchmarks you can mount already downloaded text recognition models to `/usr/local/share/ocrd-resources/` in `docker compose.yml` by adding
+To speed up QuiVer Benchmarks you can mount already downloaded text recognition models to `/usr/local/share/ocrd-resources/` in `docker-compose.yml` by adding
```yml
- path/to/your/models:/usr/local/share/ocrd-resources/
@@ -22,9 +22,12 @@ Otherwise, the tool will download all `ocrd-tesserocr-recognize` models as well
## Usage
-- clone this repository
+- clone this repository and switch to the cloned directory
- (optional) [customize](#custom-workflows-and-data) QuiVer Benchmarks according to your needs
-- run `docker compose up --build`
+- build the image with `docker compose build`
+- spin up a container with `docker compose run -d app`
+- run `docker compose exec app bash prepare.sh`
+- run `docker compose exec app bash workflows/execute_workflows.sh`
- the benchmarks and the evaluation results will be available at `data/workflows.json` on your host system
## Benchmarks Considered
diff --git a/data/workflows.json b/data/workflows.json
index aef35aa55c7f266422c4ea6df3eead048f1ba241..b377f0b605db2748574e2e5041fc832bcbc9e080 100644
--- a/data/workflows.json
+++ b/data/workflows.json
@@ -1,190 +1,108 @@
[
{
- "eval_workflow_id": "wf-data16_frak_simple_selected_pages_ocr-eval",
- "label": "Workflow on data 16_frak_simple_selected_pages_ocr",
+ "eval_workflow_id": "wf-data16_ant_complex_alberti_pictura_1540_minimal_ocr-eval",
+ "label": "Workflow on data 16_ant_complex_alberti_pictura_1540_minimal_ocr",
"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"
+ "@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/16_frak_simple.ocrd.zip",
- "label": "GT workspace 16th century Black letter simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/16_ant_complex.ocrd.zip",
+ "label": "GT workspace 16th century Antiqua complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 16_frak_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_alberti_pictura_1540_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 16_ant_complex_alberti_pictura_1540_minimal_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_frak_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_alberti_pictura_1540_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_ant_complex_alberti_pictura_1540_minimal_ocr"
},
"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": {
- "level-of-operation": "page",
- "dpi": 0,
- "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",
+ "id": "ocrd-tesserocr-recognize",
"params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
"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
+ "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": "qurator-gt4histocr-1.0",
- "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.6.2)",
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
"document_metadata": {
"data_properties": {
"fonts": [
+ "Antiqua",
"Black Letter"
],
"publication_century": "1500-1600",
"publication_decade": "",
"publication_year": "16th century",
- "number_of_pages": 6,
- "layout": "simple"
+ "number_of_pages": 3,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 249.623269,
- "cpu_time": 342.24125499999997,
- "cer_mean": 0.12233144923609811,
- "cer_median": 0.0697989946429397,
+ "wall_time": 7.386463,
+ "cpu_time": 9.83532,
+ "cer_mean": 0.10240852523716282,
+ "cer_median": 0.10536980749746708,
"cer_range": [
- 0.05172413793103448,
- 0.38848920863309355
+ 0.07124352331606218,
+ 0.1306122448979592
],
- "cer_standard_deviation": 0.13173038899014955,
- "wer": 0.26886580164226287,
- "pages_per_minute": 1.4421732454757652
+ "cer_standard_deviation": 0.02979493530847308,
+ "wer": 0.23466068901129858,
+ "pages_per_minute": 24.368902951250146
},
"by_page": [
- {
- "page_id": "phys_0001",
- "cer": 0.10179640718562874,
- "wer": 0.3008849557522124
- },
- {
- "page_id": "phys_0002",
- "cer": 0.05238095238095238,
- "wer": 0.1568627450980392
- },
- {
- "page_id": "phys_0003",
- "cer": 0.38848920863309355,
- "wer": 0.44
- },
{
"page_id": "phys_0007",
- "cer": 0.06255144032921811,
- "wer": 0.2766990291262136
+ "cer": 0.07124352331606218,
+ "wer": 0.2231404958677686
},
{
- "page_id": "phys_0021",
- "cer": 0.05172413793103448,
- "wer": 0.20833333333333334
+ "page_id": "phys_0008",
+ "cer": 0.10536980749746708,
+ "wer": 0.2484472049689441
},
{
- "page_id": "phys_0029",
- "cer": 0.07704654895666131,
- "wer": 0.2304147465437788
+ "page_id": "phys_0009",
+ "cer": 0.1306122448979592,
+ "wer": 0.2323943661971831
}
]
}
},
{
- "eval_workflow_id": "wf-data16_frak_simple_slower_processors_ocr-eval",
- "label": "Workflow on data 16_frak_simple_slower_processors_ocr",
+ "eval_workflow_id": "wf-data18_frak_simple_lessing_menschengeschlecht_1780_slower_processors_ocr-eval",
+ "label": "Workflow on data 18_frak_simple_lessing_menschengeschlecht_1780_slower_processors_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
@@ -195,16 +113,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/16_frak_simple.ocrd.zip",
- "label": "GT workspace 16th century Black letter simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_simple.ocrd.zip",
+ "label": "GT workspace 18th century Black letter simple layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 16_frak_simple_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_lessing_menschengeschlecht_1780_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_simple_lessing_menschengeschlecht_1780_slower_processors_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_frak_simple_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_lessing_menschengeschlecht_1780_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_simple_lessing_menschengeschlecht_1780_slower_processors_ocr"
},
"workflow_steps": [
{
@@ -319,157 +237,217 @@
"document_metadata": {
"data_properties": {
"fonts": [
+ "Antiqua",
"Black Letter"
],
- "publication_century": "1500-1600",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "16th century",
- "number_of_pages": 6,
+ "publication_year": "18th century",
+ "number_of_pages": 1,
"layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 93.86911700000002,
- "cpu_time": 135.81944000000001,
- "cer_mean": 0.23049280809064063,
- "cer_median": 0.09185773074661964,
+ "wall_time": 8.055356999999999,
+ "cpu_time": 12.105611,
+ "cer_mean": 0.0199501246882793,
+ "cer_median": 0.0199501246882793,
"cer_range": [
- 0.0658682634730539,
- 0.9568345323741008
+ 0.0199501246882793,
+ 0.0199501246882793
],
- "cer_standard_deviation": 0.3560945390007154,
- "wer": 0.36710708997967917,
- "pages_per_minute": 3.835127159020788
+ "cer_standard_deviation": null,
+ "wer": 0.09836065573770492,
+ "pages_per_minute": 7.44845945375233
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.0658682634730539,
- "wer": 0.18584070796460178
- },
- {
- "page_id": "phys_0002",
- "cer": 0.08412698412698413,
- "wer": 0.11764705882352941
- },
- {
- "page_id": "phys_0003",
- "cer": 0.9568345323741008,
- "wer": 0.9666666666666667
- },
- {
- "page_id": "phys_0007",
- "cer": 0.09958847736625515,
- "wer": 0.36893203883495146
- },
- {
- "page_id": "phys_0021",
- "cer": 0.10109717868338558,
- "wer": 0.2916666666666667
- },
- {
- "page_id": "phys_0029",
- "cer": 0.0754414125200642,
- "wer": 0.271889400921659
+ "cer": 0.0199501246882793,
+ "wer": 0.09836065573770492
}
]
}
},
{
- "eval_workflow_id": "wf-data19_frak_simple_minimal_ocr-eval",
- "label": "Workflow on data 19_frak_simple_minimal_ocr",
+ "eval_workflow_id": "wf-data18_frak_complex_nn_besuch_1780_selected_pages_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_nn_besuch_1780_selected_pages_ocr",
"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/19_frak_simple.ocrd.zip",
- "label": "GT workspace 19th century Black letter simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 19_frak_simple_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_nn_besuch_1780_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_nn_besuch_1780_selected_pages_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 19_frak_simple_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_nn_besuch_1780_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_nn_besuch_1780_selected_pages_ocr"
},
"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": [
- "Black Letter"
+ },
+ {
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Black Letter"
],
- "publication_century": "1800-1900",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "19th century",
- "number_of_pages": 1,
- "layout": "simple"
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 4.533288,
- "cpu_time": 5.404413,
- "cer_mean": 0.004721435316336166,
- "cer_median": 0.004721435316336166,
+ "wall_time": 76.547632,
+ "cpu_time": 113.81064699999999,
+ "cer_mean": 0.05959514724403742,
+ "cer_median": 0.0324919239726542,
"cer_range": [
- 0.004721435316336166,
- 0.004721435316336166
+ 0.008839779005524863,
+ 0.16455696202531644
],
- "cer_standard_deviation": null,
- "wer": 0.015873015873015872,
- "pages_per_minute": 13.23542647191178
+ "cer_standard_deviation": 0.07087591559854228,
+ "wer": 0.13238323035014285,
+ "pages_per_minute": 3.1353027354262246
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.004721435316336166,
- "wer": 0.015873015873015872
+ "page_id": "phys_00001",
+ "cer": 0.16455696202531644,
+ "wer": 0.27586206896551724
+ },
+ {
+ "page_id": "phys_00002",
+ "cer": 0.008839779005524863,
+ "wer": 0.04046242774566474
+ },
+ {
+ "page_id": "phys_00003",
+ "cer": 0.030501089324618737,
+ "wer": 0.09826589595375723
+ },
+ {
+ "page_id": "phys_00004",
+ "cer": 0.034482758620689655,
+ "wer": 0.11494252873563218
}
]
}
},
{
- "eval_workflow_id": "wf-data16_ant_simple_selected_pages_ocr-eval",
- "label": "Workflow on data 16_ant_simple_selected_pages_ocr",
+ "eval_workflow_id": "wf-data18_frak_complex_estor_rechtsgelehrsamkeit02_1758_selected_pages_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_estor_rechtsgelehrsamkeit02_1758_selected_pages_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
@@ -480,16 +458,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/16_ant_simple.ocrd.zip",
- "label": "GT workspace 16th century Antiqua simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 16_ant_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_estor_rechtsgelehrsamkeit02_1758_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_estor_rechtsgelehrsamkeit02_1758_selected_pages_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_ant_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_estor_rechtsgelehrsamkeit02_1758_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_estor_rechtsgelehrsamkeit02_1758_selected_pages_ocr"
},
"workflow_steps": [
{
@@ -592,52 +570,58 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua"
+ "Antiqua",
+ "Black Letter"
],
- "publication_century": "1500-1600",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "16th century",
- "number_of_pages": 3,
- "layout": "simple"
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 63.38978499999999,
- "cpu_time": 97.14254,
- "cer_mean": 0.05488709037929677,
- "cer_median": 0.03488372093023256,
+ "wall_time": 154.933488,
+ "cpu_time": 215.310644,
+ "cer_mean": 0.12432674382712249,
+ "cer_median": 0.12040572654031183,
"cer_range": [
- 0.028395061728395062,
- 0.10138248847926268
+ 0.0718294051627385,
+ 0.18466611706512778
],
- "cer_standard_deviation": 0.04039668560556835,
- "wer": 0.13745817563490756,
- "pages_per_minute": 2.8395742310847094
+ "cer_standard_deviation": 0.04665572742682036,
+ "wer": 0.18974770097390478,
+ "pages_per_minute": 1.5490518098966441
},
"by_page": [
{
- "page_id": "phys_0007",
- "cer": 0.10138248847926268,
- "wer": 0.17475728155339806
+ "page_id": "phys_0001",
+ "cer": 0.12757201646090535,
+ "wer": 0.14035087719298245
},
{
- "page_id": "phys_0013",
- "cer": 0.03488372093023256,
- "wer": 0.13592233009708737
+ "page_id": "phys_0002",
+ "cer": 0.1132394366197183,
+ "wer": 0.1950354609929078
},
{
- "page_id": "phys_0014",
- "cer": 0.028395061728395062,
- "wer": 0.1016949152542373
+ "page_id": "phys_0003",
+ "cer": 0.0718294051627385,
+ "wer": 0.15087719298245614
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.18466611706512778,
+ "wer": 0.2727272727272727
}
]
}
},
{
- "eval_workflow_id": "wf-data16_ant_simple_slower_processors_ocr-eval",
- "label": "Workflow on data 16_ant_simple_slower_processors_ocr",
+ "eval_workflow_id": "wf-data18_fontmix_complex_benner_herrnhuterey04_1748_slower_processors_ocr-eval",
+ "label": "Workflow on data 18_fontmix_complex_benner_herrnhuterey04_1748_slower_processors_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
@@ -648,16 +632,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/16_ant_simple.ocrd.zip",
- "label": "GT workspace 16th century Antiqua simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_fontmix_complex.ocrd.zip",
+ "label": "GT workspace 18th century Font Mix complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 16_ant_simple_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_benner_herrnhuterey04_1748_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 18_fontmix_complex_benner_herrnhuterey04_1748_slower_processors_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_ant_simple_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_benner_herrnhuterey04_1748_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_fontmix_complex_benner_herrnhuterey04_1748_slower_processors_ocr"
},
"workflow_steps": [
{
@@ -772,274 +756,90 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua"
+ "Antiqua",
+ "Black Letter"
],
- "publication_century": "1500-1600",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "16th century",
- "number_of_pages": 3,
- "layout": "simple"
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 28.936798,
- "cpu_time": 46.368261,
- "cer_mean": 0.06721136853878373,
- "cer_median": 0.055232558139534885,
+ "wall_time": 67.309474,
+ "cpu_time": 101.85330800000001,
+ "cer_mean": 0.16726583056278752,
+ "cer_median": 0.09637318392327315,
"cer_range": [
- 0.03580246913580247,
- 0.11059907834101383
+ 0.03187250996015936,
+ 0.4444444444444444
],
- "cer_standard_deviation": 0.038810463938030185,
- "wer": 0.1847677033624047,
- "pages_per_minute": 6.220453278901141
+ "cer_standard_deviation": 0.18889822286887584,
+ "wer": 0.28912998545359864,
+ "pages_per_minute": 3.565619900699269
},
"by_page": [
{
- "page_id": "phys_0007",
- "cer": 0.11059907834101383,
- "wer": 0.23300970873786409
+ "page_id": "phys_0001",
+ "cer": 0.4444444444444444,
+ "wer": 0.6204379562043796
},
{
- "page_id": "phys_0013",
- "cer": 0.055232558139534885,
- "wer": 0.1941747572815534
+ "page_id": "phys_0002",
+ "cer": 0.03187250996015936,
+ "wer": 0.10919540229885058
},
{
- "page_id": "phys_0014",
- "cer": 0.03580246913580247,
- "wer": 0.1271186440677966
+ "page_id": "phys_0003",
+ "cer": 0.1266778523489933,
+ "wer": 0.2681564245810056
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.06606851549755302,
+ "wer": 0.15873015873015872
}
]
}
},
{
- "eval_workflow_id": "wf-data18_frak_simple_minimal_ocr-eval",
- "label": "Workflow on data 18_frak_simple_minimal_ocr",
+ "eval_workflow_id": "wf-data18_frak_complex_justi_abhandlung01_1758_selected_pages_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_justi_abhandlung01_1758_selected_pages_ocr",
"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/18_frak_simple.ocrd.zip",
- "label": "GT workspace 18th century Black letter simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 18_frak_simple_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_justi_abhandlung01_1758_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_justi_abhandlung01_1758_selected_pages_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_frak_simple_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_justi_abhandlung01_1758_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_justi_abhandlung01_1758_selected_pages_ocr"
},
"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"
- }
- }
- ],
- "workflow_model": "Fraktur_GT4HistOCR",
- "eval_tool": "ocrd-dinglehopper vNone",
- "document_metadata": {
- "data_properties": {
- "fonts": [
- "Black Letter"
- ],
- "publication_century": "1700-1800",
- "publication_decade": "",
- "publication_year": "18th century",
- "number_of_pages": 1,
- "layout": "simple"
- }
- }
- },
- "evaluation_results": {
- "document_wide": {
- "wall_time": 1.900465,
- "cpu_time": 2.519007,
- "cer_mean": 0.02493765586034913,
- "cer_median": 0.02493765586034913,
- "cer_range": [
- 0.02493765586034913,
- 0.02493765586034913
- ],
- "cer_standard_deviation": null,
- "wer": 0.09836065573770492,
- "pages_per_minute": 31.571220727558778
- },
- "by_page": [
- {
- "page_id": "phys_0001",
- "cer": 0.02493765586034913,
- "wer": 0.09836065573770492
- }
- ]
- }
- },
- {
- "eval_workflow_id": "wf-data19_ant_simple_minimal_ocr-eval",
- "label": "Workflow on data 19_ant_simple_minimal_ocr",
- "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/19_ant_simple.ocrd.zip",
- "label": "GT workspace 19th century Antiqua simple layout"
- },
- "ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 19_ant_simple_minimal_ocr"
- },
- "eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 19_ant_simple_minimal_ocr"
- },
- "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": "1800-1900",
- "publication_decade": "",
- "publication_year": "19th century",
- "number_of_pages": 3,
- "layout": "simple"
- }
- }
- },
- "evaluation_results": {
- "document_wide": {
- "wall_time": 12.639051,
- "cpu_time": 15.064324,
- "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": 14.241575573988902
- },
- "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-data19_frak_simple_slower_processors_ocr-eval",
- "label": "Workflow on data 19_frak_simple_slower_processors_ocr",
- "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/19_frak_simple.ocrd.zip",
- "label": "GT workspace 19th century Black letter simple layout"
- },
- "ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 19_frak_simple_slower_processors_ocr"
- },
- "eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 19_frak_simple_slower_processors_ocr"
- },
- "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"
+ "level-of-operation": "page"
}
},
{
@@ -1061,11 +861,22 @@
}
},
{
- "id": "ocrd-cis-ocropy-denoise",
+ "id": "ocrd-skimage-binarize",
"params": {
+ "method": "li",
"level-of-operation": "page",
- "noise_maxsize": 3.0,
- "dpi": 0
+ "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
}
},
{
@@ -1077,28 +888,22 @@
}
},
{
- "id": "ocrd-tesserocr-segment",
+ "id": "ocrd-cis-ocropy-segment",
"params": {
+ "level-of-operation": "page",
"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"
+ "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
}
},
{
@@ -1111,73 +916,72 @@
}
},
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-calamari-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"
+ "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 2.6.2)",
"document_metadata": {
"data_properties": {
"fonts": [
+ "Antiqua",
"Black Letter"
],
- "publication_century": "1800-1900",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "19th century",
- "number_of_pages": 1,
- "layout": "simple"
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 13.078715,
- "cpu_time": 20.989223000000003,
- "cer_mean": 0.0056657223796034,
- "cer_median": 0.0056657223796034,
+ "wall_time": 122.82167999999999,
+ "cpu_time": 173.67148600000002,
+ "cer_mean": 0.05991629368071161,
+ "cer_median": 0.0496849827491332,
"cer_range": [
- 0.0056657223796034,
- 0.0056657223796034
+ 0.0019973368841544607,
+ 0.13829787234042554
],
- "cer_standard_deviation": null,
- "wer": 0.031746031746031744,
- "pages_per_minute": 4.58760665707602
+ "cer_standard_deviation": 0.06406814571829976,
+ "wer": 0.09724037283059882,
+ "pages_per_minute": 1.954052411593784
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.0056657223796034,
- "wer": 0.031746031746031744
+ "cer": 0.013584117032392894,
+ "wer": 0.06666666666666667
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.13829787234042554,
+ "wer": 0.1875
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.08578584846587352,
+ "wer": 0.12598425196850394
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.0019973368841544607,
+ "wer": 0.00881057268722467
}
]
}
},
{
- "eval_workflow_id": "wf-data18_frak_simple_selected_pages_ocr-eval",
- "label": "Workflow on data 18_frak_simple_selected_pages_ocr",
+ "eval_workflow_id": "wf-data18_frak_complex_luz_blitz_1784_selected_pages_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_luz_blitz_1784_selected_pages_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
@@ -1188,16 +992,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_simple.ocrd.zip",
- "label": "GT workspace 18th century Black letter simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 18_frak_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_luz_blitz_1784_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_luz_blitz_1784_selected_pages_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_frak_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_luz_blitz_1784_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_luz_blitz_1784_selected_pages_ocr"
},
"workflow_steps": [
{
@@ -1300,42 +1104,164 @@
"document_metadata": {
"data_properties": {
"fonts": [
+ "Antiqua",
"Black Letter"
],
"publication_century": "1700-1800",
"publication_decade": "",
"publication_year": "18th century",
- "number_of_pages": 1,
- "layout": "simple"
+ "number_of_pages": 4,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 24.445896000000005,
- "cpu_time": 35.454705999999995,
- "cer_mean": 0.02493765586034913,
- "cer_median": 0.02493765586034913,
+ "wall_time": 145.735792,
+ "cpu_time": 208.46528999999998,
+ "cer_mean": 0.05256131079939043,
+ "cer_median": 0.029898228835687546,
"cer_range": [
- 0.02493765586034913,
- 0.02493765586034913
+ 0.009184845005740528,
+ 0.1412639405204461
],
- "cer_standard_deviation": null,
- "wer": 0.09836065573770492,
- "pages_per_minute": 2.4543997078282582
+ "cer_standard_deviation": 0.06044359233653714,
+ "wer": 0.15936235092429332,
+ "pages_per_minute": 1.6468157664384877
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.02493765586034913,
- "wer": 0.09836065573770492
- }
- ]
+ "cer": 0.1412639405204461,
+ "wer": 0.38461538461538464
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.039473684210526314,
+ "wer": 0.13043478260869565
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.020322773460848775,
+ "wer": 0.06792452830188679
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.009184845005740528,
+ "wer": 0.054474708171206226
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data18_frak_complex_luz_blitz_1784_minimal_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_luz_blitz_1784_minimal_ocr",
+ "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/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_luz_blitz_1784_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_luz_blitz_1784_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_luz_blitz_1784_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_luz_blitz_1784_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 18.435259,
+ "cpu_time": 21.704126,
+ "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": 13.018531499882915
+ },
+ "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-data18_frak_simple_slower_processors_ocr-eval",
- "label": "Workflow on data 18_frak_simple_slower_processors_ocr",
+ "eval_workflow_id": "wf-data16_frak_simple_luther_auszlegunge_1520_slower_processors_ocr-eval",
+ "label": "Workflow on data 16_frak_simple_luther_auszlegunge_1520_slower_processors_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
@@ -1346,16 +1272,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_simple.ocrd.zip",
- "label": "GT workspace 18th century Black letter simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/16_frak_simple.ocrd.zip",
+ "label": "GT workspace 16th century Black letter simple layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 18_frak_simple_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_luther_auszlegunge_1520_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 16_frak_simple_luther_auszlegunge_1520_slower_processors_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_frak_simple_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_luther_auszlegunge_1520_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_frak_simple_luther_auszlegunge_1520_slower_processors_ocr"
},
"workflow_steps": [
{
@@ -1470,62 +1396,68 @@
"document_metadata": {
"data_properties": {
"fonts": [
+ "Antiqua",
"Black Letter"
],
- "publication_century": "1700-1800",
+ "publication_century": "1500-1600",
"publication_decade": "",
- "publication_year": "18th century",
- "number_of_pages": 1,
+ "publication_year": "16th century",
+ "number_of_pages": 2,
"layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 7.974276000000001,
- "cpu_time": 11.831112000000001,
- "cer_mean": 0.0199501246882793,
- "cer_median": 0.0199501246882793,
+ "wall_time": 36.087631,
+ "cpu_time": 47.946267999999996,
+ "cer_mean": 0.5161379724470825,
+ "cer_median": 0.5161379724470825,
"cer_range": [
- 0.0199501246882793,
- 0.0199501246882793
+ 0.0754414125200642,
+ 0.9568345323741008
],
- "cer_standard_deviation": null,
- "wer": 0.09836065573770492,
- "pages_per_minute": 7.524194045954767
+ "cer_standard_deviation": 0.6232390519399567,
+ "wer": 0.6192780337941628,
+ "pages_per_minute": 3.3252390548994475
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.0199501246882793,
- "wer": 0.09836065573770492
+ "page_id": "phys_0003",
+ "cer": 0.9568345323741008,
+ "wer": 0.9666666666666667
+ },
+ {
+ "page_id": "phys_0029",
+ "cer": 0.0754414125200642,
+ "wer": 0.271889400921659
}
]
}
},
{
- "eval_workflow_id": "wf-data18_fontmix_complex_selected_pages_ocr-eval",
- "label": "Workflow on data 18_fontmix_complex_selected_pages_ocr",
+ "eval_workflow_id": "wf-data18_frak_complex_bernd_lebensbeschreibung_1738_slower_processors_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_bernd_lebensbeschreibung_1738_slower_processors_ocr",
"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"
+ "@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/18_fontmix_complex.ocrd.zip",
- "label": "GT workspace 18th century Font Mix complex layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 18_fontmix_complex_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_bernd_lebensbeschreibung_1738_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_bernd_lebensbeschreibung_1738_slower_processors_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_fontmix_complex_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_bernd_lebensbeschreibung_1738_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_bernd_lebensbeschreibung_1738_slower_processors_ocr"
},
"workflow_steps": [
{
@@ -1559,22 +1491,11 @@
}
},
{
- "id": "ocrd-skimage-binarize",
- "params": {
- "method": "li",
- "level-of-operation": "page",
- "dpi": 0,
- "window_size": 301,
- "k": 0.34
- }
- },
- {
- "id": "ocrd-skimage-denoise",
+ "id": "ocrd-cis-ocropy-denoise",
"params": {
"level-of-operation": "page",
- "dpi": 0,
- "protect": 0.0,
- "maxsize": 1.0
+ "noise_maxsize": 3.0,
+ "dpi": 0
}
},
{
@@ -1586,22 +1507,28 @@
}
},
{
- "id": "ocrd-cis-ocropy-segment",
+ "id": "ocrd-tesserocr-segment",
"params": {
- "level-of-operation": "page",
"dpi": 0,
- "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
+ "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"
}
},
{
@@ -1614,17 +1541,34 @@
}
},
{
- "id": "ocrd-calamari-recognize",
+ "id": "ocrd-tesserocr-recognize",
"params": {
- "checkpoint_dir": "qurator-gt4histocr-1.0",
- "voter": "confidence_voter_default_ctc",
- "textequiv_level": "line",
- "glyph_conf_cutoff": 0.001
+ "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": "qurator-gt4histocr-1.0",
- "eval_tool": "ocrd-calamari-recognize v1.0.5 (calamari 1.0.5, tensorflow 2.6.2)",
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
"document_metadata": {
"data_properties": {
"fonts": [
@@ -1634,94 +1578,5365 @@
"publication_century": "1700-1800",
"publication_decade": "",
"publication_year": "18th century",
- "number_of_pages": 4,
+ "number_of_pages": 3,
"layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 155.643285,
- "cpu_time": 267.033236,
- "cer_mean": 0.140061110527823,
- "cer_median": 0.06466848046246318,
+ "wall_time": 34.033219,
+ "cpu_time": 53.635996999999996,
+ "cer_mean": 0.06688670097710152,
+ "cer_median": 0.03746177370030581,
"cer_range": [
- 0.028685258964143426,
- 0.4022222222222222
+ 0.0290519877675841,
+ 0.13414634146341464
],
- "cer_standard_deviation": 0.17581800739756628,
- "wer": 0.23651027895776314,
- "pages_per_minute": 1.541987500456573
+ "cer_standard_deviation": 0.058400133164399716,
+ "wer": 0.12383493220296966,
+ "pages_per_minute": 5.2889501871686
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.4022222222222222,
- "wer": 0.5474452554744526
+ "cer": 0.13414634146341464,
+ "wer": 0.22058823529411764
},
{
"page_id": "phys_0002",
- "cer": 0.028685258964143426,
- "wer": 0.09195402298850575
+ "cer": 0.0290519877675841,
+ "wer": 0.05357142857142857
},
{
"page_id": "phys_0003",
- "cer": 0.07550335570469799,
- "wer": 0.19553072625698323
- },
- {
- "page_id": "phys_0004",
- "cer": 0.053833605220228384,
- "wer": 0.1111111111111111
+ "cer": 0.03746177370030581,
+ "wer": 0.09734513274336283
}
]
}
},
{
- "eval_workflow_id": "wf-data16_frak_simple_minimal_ocr-eval",
- "label": "Workflow on data 16_frak_simple_minimal_ocr",
+ "eval_workflow_id": "wf-data18_frak_complex_euler_rechenkunst01_1738_slower_processors_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_euler_rechenkunst01_1738_slower_processors_ocr",
"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/16_frak_simple.ocrd.zip",
- "label": "GT workspace 16th century Black letter simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 16_frak_simple_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_euler_rechenkunst01_1738_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_euler_rechenkunst01_1738_slower_processors_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_frak_simple_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_euler_rechenkunst01_1738_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_euler_rechenkunst01_1738_slower_processors_ocr"
},
"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": {},
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 6,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 78.308453,
+ "cpu_time": 124.319609,
+ "cer_mean": 0.1500034071140843,
+ "cer_median": 0.12202020202020203,
+ "cer_range": [
+ 0.03902862098872507,
+ 0.39473684210526316
+ ],
+ "cer_standard_deviation": 0.12934283380557174,
+ "wer": 0.27891577144281926,
+ "pages_per_minute": 4.597204850924586
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.03902862098872507,
+ "wer": 0.0872093023255814
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.39473684210526316,
+ "wer": 0.6402439024390244
+ },
+ {
+ "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-data16_frak_simple_luther_auszlegunge_1520_minimal_ocr-eval",
+ "label": "Workflow on data 16_frak_simple_luther_auszlegunge_1520_minimal_ocr",
+ "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/16_frak_simple.ocrd.zip",
+ "label": "GT workspace 16th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_luther_auszlegunge_1520_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 16_frak_simple_luther_auszlegunge_1520_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_luther_auszlegunge_1520_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_frak_simple_luther_auszlegunge_1520_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 10.980085,
+ "cpu_time": 12.980711,
+ "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": 10.928877144393692
+ },
+ "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-data18_frak_complex_buerger_gedichte_1778_slower_processors_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_buerger_gedichte_1778_slower_processors_ocr",
+ "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/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_buerger_gedichte_1778_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_buerger_gedichte_1778_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_buerger_gedichte_1778_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_buerger_gedichte_1778_slower_processors_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 2,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 16.86402,
+ "cpu_time": 26.750532,
+ "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": 7.115741086644821
+ },
+ "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-data19_frak_simple_arnimb_goethe03_1835_selected_pages_ocr-eval",
+ "label": "Workflow on data 19_frak_simple_arnimb_goethe03_1835_selected_pages_ocr",
+ "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/19_frak_simple.ocrd.zip",
+ "label": "GT workspace 19th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_arnimb_goethe03_1835_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 19_frak_simple_arnimb_goethe03_1835_selected_pages_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_arnimb_goethe03_1835_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 19_frak_simple_arnimb_goethe03_1835_selected_pages_ocr"
+ },
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Black Letter"
+ ],
+ "publication_century": "1800-1900",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 1,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 45.820398,
+ "cpu_time": 66.91274899999999,
+ "cer_mean": 0.007554296506137866,
+ "cer_median": 0.007554296506137866,
+ "cer_range": [
+ 0.007554296506137866,
+ 0.007554296506137866
+ ],
+ "cer_standard_deviation": null,
+ "wer": 0.015873015873015872,
+ "pages_per_minute": 1.309460472167876
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.007554296506137866,
+ "wer": 0.015873015873015872
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data16_ant_simple_heyden_paedono_1548_slower_processors_ocr-eval",
+ "label": "Workflow on data 16_ant_simple_heyden_paedono_1548_slower_processors_ocr",
+ "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/16_ant_simple.ocrd.zip",
+ "label": "GT workspace 16th century Antiqua simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_heyden_paedono_1548_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 16_ant_simple_heyden_paedono_1548_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_heyden_paedono_1548_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_ant_simple_heyden_paedono_1548_slower_processors_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 32.539348,
+ "cpu_time": 51.522006000000005,
+ "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": 5.5317641890058775
+ },
+ "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-data16_ant_complex_alberti_pictura_1540_slower_processors_ocr-eval",
+ "label": "Workflow on data 16_ant_complex_alberti_pictura_1540_slower_processors_ocr",
+ "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/16_ant_complex.ocrd.zip",
+ "label": "GT workspace 16th century Antiqua complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_alberti_pictura_1540_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 16_ant_complex_alberti_pictura_1540_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_alberti_pictura_1540_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_ant_complex_alberti_pictura_1540_slower_processors_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 3,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 30.780167000000002,
+ "cpu_time": 52.275692,
+ "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": 5.8479214878853645
+ },
+ "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-data19_frak_simple_arnimb_goethe03_1835_minimal_ocr-eval",
+ "label": "Workflow on data 19_frak_simple_arnimb_goethe03_1835_minimal_ocr",
+ "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/19_frak_simple.ocrd.zip",
+ "label": "GT workspace 19th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_arnimb_goethe03_1835_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 19_frak_simple_arnimb_goethe03_1835_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_arnimb_goethe03_1835_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 19_frak_simple_arnimb_goethe03_1835_minimal_ocr"
+ },
+ "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": [
+ "Black Letter"
+ ],
+ "publication_century": "1800-1900",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 1,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 4.437351,
+ "cpu_time": 5.324579,
+ "cer_mean": 0.004721435316336166,
+ "cer_median": 0.004721435316336166,
+ "cer_range": [
+ 0.004721435316336166,
+ 0.004721435316336166
+ ],
+ "cer_standard_deviation": null,
+ "wer": 0.015873015873015872,
+ "pages_per_minute": 13.521580780965943
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.004721435316336166,
+ "wer": 0.015873015873015872
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data18_fontmix_complex_benner_herrnhuterey04_1748_minimal_ocr-eval",
+ "label": "Workflow on data 18_fontmix_complex_benner_herrnhuterey04_1748_minimal_ocr",
+ "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/18_fontmix_complex.ocrd.zip",
+ "label": "GT workspace 18th century Font Mix complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_benner_herrnhuterey04_1748_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 18_fontmix_complex_benner_herrnhuterey04_1748_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_benner_herrnhuterey04_1748_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_fontmix_complex_benner_herrnhuterey04_1748_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 20.134204,
+ "cpu_time": 23.74436,
+ "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": 11.920014319910536
+ },
+ "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-data16_ant_simple_heyden_paedono_1548_minimal_ocr-eval",
+ "label": "Workflow on data 16_ant_simple_heyden_paedono_1548_minimal_ocr",
+ "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/16_ant_simple.ocrd.zip",
+ "label": "GT workspace 16th century Antiqua simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_heyden_paedono_1548_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 16_ant_simple_heyden_paedono_1548_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_heyden_paedono_1548_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_ant_simple_heyden_paedono_1548_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 8.030865,
+ "cpu_time": 10.331791,
+ "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": 22.41352581571225
+ },
+ "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-data18_ant_simple_ballenstedt_delatio_1777_slower_processors_ocr-eval",
+ "label": "Workflow on data 18_ant_simple_ballenstedt_delatio_1777_slower_processors_ocr",
+ "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/18_ant_simple.ocrd.zip",
+ "label": "GT workspace 18th century Antiqua simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_ant_simple_ballenstedt_delatio_1777_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 18_ant_simple_ballenstedt_delatio_1777_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_ant_simple_ballenstedt_delatio_1777_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_ant_simple_ballenstedt_delatio_1777_slower_processors_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 73.616766,
+ "cpu_time": 113.02993400000001,
+ "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": 2.445095183887866
+ },
+ "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-data16_frak_simple_kistler_kraeuter_1500_slower_processors_ocr-eval",
+ "label": "Workflow on data 16_frak_simple_kistler_kraeuter_1500_slower_processors_ocr",
+ "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/16_frak_simple.ocrd.zip",
+ "label": "GT workspace 16th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_kistler_kraeuter_1500_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 16_frak_simple_kistler_kraeuter_1500_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_kistler_kraeuter_1500_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_frak_simple_kistler_kraeuter_1500_slower_processors_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 34.966049000000005,
+ "cpu_time": 54.605062,
+ "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": 3.4319004700817066
+ },
+ "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-data18_frak_complex_bernd_lebensbeschreibung_1738_minimal_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_bernd_lebensbeschreibung_1738_minimal_ocr",
+ "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/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_bernd_lebensbeschreibung_1738_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_bernd_lebensbeschreibung_1738_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_bernd_lebensbeschreibung_1738_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_bernd_lebensbeschreibung_1738_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 3,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 8.242131,
+ "cpu_time": 10.755872,
+ "cer_mean": 0.1790109644215708,
+ "cer_median": 0.05504587155963303,
+ "cer_range": [
+ 0.008409785932721712,
+ 0.4735772357723577
+ ],
+ "cer_standard_deviation": 0.2561653709691831,
+ "wer": 0.2391410103864555,
+ "pages_per_minute": 21.83901226515327
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.4735772357723577,
+ "wer": 0.5882352941176471
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.05504587155963303,
+ "wer": 0.09821428571428571
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.008409785932721712,
+ "wer": 0.030973451327433628
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data19_ant_simple_blumenbach_anatomie_1805_slower_processors_ocr-eval",
+ "label": "Workflow on data 19_ant_simple_blumenbach_anatomie_1805_slower_processors_ocr",
+ "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/19_ant_simple.ocrd.zip",
+ "label": "GT workspace 19th century Antiqua simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_blumenbach_anatomie_1805_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 19_ant_simple_blumenbach_anatomie_1805_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_blumenbach_anatomie_1805_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 19_ant_simple_blumenbach_anatomie_1805_slower_processors_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1800-1900",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 42.362522,
+ "cpu_time": 65.91597100000001,
+ "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": 4.249038808407111
+ },
+ "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-data18_frak_complex_justi_abhandlung01_1758_slower_processors_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_justi_abhandlung01_1758_slower_processors_ocr",
+ "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/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_justi_abhandlung01_1758_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_justi_abhandlung01_1758_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_justi_abhandlung01_1758_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_justi_abhandlung01_1758_slower_processors_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 55.553484,
+ "cpu_time": 89.664067,
+ "cer_mean": 0.06272817755704542,
+ "cer_median": 0.04686327354908531,
+ "cer_range": [
+ 0.0039946737683089215,
+ 0.15319148936170213
+ ],
+ "cer_standard_deviation": 0.07133216362479361,
+ "wer": 0.0881207064150021,
+ "pages_per_minute": 4.320161090166731
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.0073145245559038665,
+ "wer": 0.02857142857142857
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.15319148936170213,
+ "wer": 0.20833333333333334
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.08641202254226675,
+ "wer": 0.10236220472440945
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.0039946737683089215,
+ "wer": 0.013215859030837005
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data18_fontmix_complex_benner_herrnhuterey04_1748_selected_pages_ocr-eval",
+ "label": "Workflow on data 18_fontmix_complex_benner_herrnhuterey04_1748_selected_pages_ocr",
+ "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/18_fontmix_complex.ocrd.zip",
+ "label": "GT workspace 18th century Font Mix complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_benner_herrnhuterey04_1748_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 18_fontmix_complex_benner_herrnhuterey04_1748_selected_pages_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_benner_herrnhuterey04_1748_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_fontmix_complex_benner_herrnhuterey04_1748_selected_pages_ocr"
+ },
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 151.101655,
+ "cpu_time": 259.05950800000005,
+ "cer_mean": 0.140061110527823,
+ "cer_median": 0.06466848046246318,
+ "cer_range": [
+ 0.028685258964143426,
+ 0.4022222222222222
+ ],
+ "cer_standard_deviation": 0.17581800739756628,
+ "wer": 0.23651027895776314,
+ "pages_per_minute": 1.5883346876644073
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.4022222222222222,
+ "wer": 0.5474452554744526
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.028685258964143426,
+ "wer": 0.09195402298850575
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.07550335570469799,
+ "wer": 0.19553072625698323
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.053833605220228384,
+ "wer": 0.1111111111111111
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data18_frak_simple_lessing_menschengeschlecht_1780_minimal_ocr-eval",
+ "label": "Workflow on data 18_frak_simple_lessing_menschengeschlecht_1780_minimal_ocr",
+ "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/18_frak_simple.ocrd.zip",
+ "label": "GT workspace 18th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_lessing_menschengeschlecht_1780_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_simple_lessing_menschengeschlecht_1780_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_lessing_menschengeschlecht_1780_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_simple_lessing_menschengeschlecht_1780_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 1,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 1.751117,
+ "cpu_time": 2.32017,
+ "cer_mean": 0.02493765586034913,
+ "cer_median": 0.02493765586034913,
+ "cer_range": [
+ 0.02493765586034913,
+ 0.02493765586034913
+ ],
+ "cer_standard_deviation": null,
+ "wer": 0.09836065573770492,
+ "pages_per_minute": 34.263844163468235
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.02493765586034913,
+ "wer": 0.09836065573770492
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data16_frak_simple_trota_mordtbrenner_1540_slower_processors_ocr-eval",
+ "label": "Workflow on data 16_frak_simple_trota_mordtbrenner_1540_slower_processors_ocr",
+ "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/16_frak_simple.ocrd.zip",
+ "label": "GT workspace 16th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_trota_mordtbrenner_1540_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 16_frak_simple_trota_mordtbrenner_1540_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_trota_mordtbrenner_1540_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_frak_simple_trota_mordtbrenner_1540_slower_processors_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 28.966472,
+ "cpu_time": 40.91749,
+ "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": 4.14272059089557
+ },
+ "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-data16_frak_simple_trota_mordtbrenner_1540_selected_pages_ocr-eval",
+ "label": "Workflow on data 16_frak_simple_trota_mordtbrenner_1540_selected_pages_ocr",
+ "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/16_frak_simple.ocrd.zip",
+ "label": "GT workspace 16th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_trota_mordtbrenner_1540_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 16_frak_simple_trota_mordtbrenner_1540_selected_pages_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_trota_mordtbrenner_1540_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_frak_simple_trota_mordtbrenner_1540_selected_pages_ocr"
+ },
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 108.32675099999999,
+ "cpu_time": 136.94418900000002,
+ "cer_mean": 0.07708867978329056,
+ "cer_median": 0.07708867978329056,
+ "cer_range": [
+ 0.05238095238095238,
+ 0.10179640718562874
+ ],
+ "cer_standard_deviation": 0.034942003187804015,
+ "wer": 0.2288738504251258,
+ "pages_per_minute": 1.1077596151665254
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.10179640718562874,
+ "wer": 0.3008849557522124
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.05238095238095238,
+ "wer": 0.1568627450980392
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data18_frak_simple_lessing_menschengeschlecht_1780_selected_pages_ocr-eval",
+ "label": "Workflow on data 18_frak_simple_lessing_menschengeschlecht_1780_selected_pages_ocr",
+ "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/18_frak_simple.ocrd.zip",
+ "label": "GT workspace 18th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_lessing_menschengeschlecht_1780_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_simple_lessing_menschengeschlecht_1780_selected_pages_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_simple_lessing_menschengeschlecht_1780_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_simple_lessing_menschengeschlecht_1780_selected_pages_ocr"
+ },
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 1,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 23.314418,
+ "cpu_time": 34.297031000000004,
+ "cer_mean": 0.02493765586034913,
+ "cer_median": 0.02493765586034913,
+ "cer_range": [
+ 0.02493765586034913,
+ 0.02493765586034913
+ ],
+ "cer_standard_deviation": null,
+ "wer": 0.09836065573770492,
+ "pages_per_minute": 2.5735148095912153
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.02493765586034913,
+ "wer": 0.09836065573770492
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data16_frak_simple_luther_auszlegunge_1520_selected_pages_ocr-eval",
+ "label": "Workflow on data 16_frak_simple_luther_auszlegunge_1520_selected_pages_ocr",
+ "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/16_frak_simple.ocrd.zip",
+ "label": "GT workspace 16th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_luther_auszlegunge_1520_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 16_frak_simple_luther_auszlegunge_1520_selected_pages_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_luther_auszlegunge_1520_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_frak_simple_luther_auszlegunge_1520_selected_pages_ocr"
+ },
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 106.636709,
+ "cpu_time": 141.711402,
+ "cer_mean": 0.23276787879487743,
+ "cer_median": 0.23276787879487743,
+ "cer_range": [
+ 0.07704654895666131,
+ 0.38848920863309355
+ ],
+ "cer_standard_deviation": 0.22022321660797936,
+ "wer": 0.33520737327188943,
+ "pages_per_minute": 1.125316048528842
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0003",
+ "cer": 0.38848920863309355,
+ "wer": 0.44
+ },
+ {
+ "page_id": "phys_0029",
+ "cer": 0.07704654895666131,
+ "wer": 0.2304147465437788
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data17_frak_simple_calvi_beutelschneider01_1627_minimal_ocr-eval",
+ "label": "Workflow on data 17_frak_simple_calvi_beutelschneider01_1627_minimal_ocr",
+ "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/17_frak_simple.ocrd.zip",
+ "label": "GT workspace 17th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_calvi_beutelschneider01_1627_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 17_frak_simple_calvi_beutelschneider01_1627_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_calvi_beutelschneider01_1627_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 17_frak_simple_calvi_beutelschneider01_1627_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 7.886743,
+ "cpu_time": 10.460914,
+ "cer_mean": 0.08427970798416445,
+ "cer_median": 0.09331797235023041,
+ "cer_range": [
+ 0.05025996533795494,
+ 0.10926118626430802
+ ],
+ "cer_standard_deviation": 0.030521364398782277,
+ "wer": 0.1842438386542771,
+ "pages_per_minute": 22.82310961571843
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.10926118626430802,
+ "wer": 0.2229299363057325
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.09331797235023041,
+ "wer": 0.20689655172413793
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.05025996533795494,
+ "wer": 0.12290502793296089
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data19_ant_simple_blumenbach_anatomie_1805_minimal_ocr-eval",
+ "label": "Workflow on data 19_ant_simple_blumenbach_anatomie_1805_minimal_ocr",
+ "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/19_ant_simple.ocrd.zip",
+ "label": "GT workspace 19th century Antiqua simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_blumenbach_anatomie_1805_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 19_ant_simple_blumenbach_anatomie_1805_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_blumenbach_anatomie_1805_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 19_ant_simple_blumenbach_anatomie_1805_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1800-1900",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 12.28958,
+ "cpu_time": 14.798878,
+ "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": 14.646554235376634
+ },
+ "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-data18_frak_complex_nn_besuch_1780_minimal_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_nn_besuch_1780_minimal_ocr",
+ "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/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_nn_besuch_1780_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_nn_besuch_1780_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_nn_besuch_1780_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_nn_besuch_1780_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 13.367254,
+ "cpu_time": 16.145112,
+ "cer_mean": 0.06315407734401027,
+ "cer_median": 0.026726016076928857,
+ "cer_range": [
+ 0.01878453038674033,
+ 0.18037974683544303
+ ],
+ "cer_standard_deviation": 0.07825196427362012,
+ "wer": 0.11662846322503488,
+ "pages_per_minute": 17.95432330379897
+ },
+ "by_page": [
+ {
+ "page_id": "phys_00001",
+ "cer": 0.18037974683544303,
+ "wer": 0.1896551724137931
+ },
+ {
+ "page_id": "phys_00002",
+ "cer": 0.01878453038674033,
+ "wer": 0.08670520231213873
+ },
+ {
+ "page_id": "phys_00003",
+ "cer": 0.02505446623093682,
+ "wer": 0.08670520231213873
+ },
+ {
+ "page_id": "phys_00004",
+ "cer": 0.028397565922920892,
+ "wer": 0.10344827586206896
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data16_frak_simple_kistler_kraeuter_1500_minimal_ocr-eval",
+ "label": "Workflow on data 16_frak_simple_kistler_kraeuter_1500_minimal_ocr",
+ "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/16_frak_simple.ocrd.zip",
+ "label": "GT workspace 16th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_kistler_kraeuter_1500_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 16_frak_simple_kistler_kraeuter_1500_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_kistler_kraeuter_1500_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_frak_simple_kistler_kraeuter_1500_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 8.741153,
+ "cpu_time": 10.631666,
+ "cer_mean": 0.11307713146793608,
+ "cer_median": 0.11307713146793608,
+ "cer_range": [
+ 0.09135802469135802,
+ 0.13479623824451412
+ ],
+ "cer_standard_deviation": 0.03071545536606607,
+ "wer": 0.35497572815533984,
+ "pages_per_minute": 13.728166066879277
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0007",
+ "cer": 0.09135802469135802,
+ "wer": 0.33495145631067963
+ },
+ {
+ "page_id": "phys_0021",
+ "cer": 0.13479623824451412,
+ "wer": 0.375
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data16_frak_simple_kistler_kraeuter_1500_selected_pages_ocr-eval",
+ "label": "Workflow on data 16_frak_simple_kistler_kraeuter_1500_selected_pages_ocr",
+ "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/16_frak_simple.ocrd.zip",
+ "label": "GT workspace 16th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_kistler_kraeuter_1500_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 16_frak_simple_kistler_kraeuter_1500_selected_pages_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_kistler_kraeuter_1500_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_frak_simple_kistler_kraeuter_1500_selected_pages_ocr"
+ },
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 81.76006000000001,
+ "cpu_time": 115.33805,
+ "cer_mean": 0.05713778913012629,
+ "cer_median": 0.05713778913012629,
+ "cer_range": [
+ 0.05172413793103448,
+ 0.06255144032921811
+ ],
+ "cer_standard_deviation": 0.0076560589477130125,
+ "wer": 0.24251618122977348,
+ "pages_per_minute": 1.4677092947338832
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0007",
+ "cer": 0.06255144032921811,
+ "wer": 0.2766990291262136
+ },
+ {
+ "page_id": "phys_0021",
+ "cer": 0.05172413793103448,
+ "wer": 0.20833333333333334
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data18_frak_complex_estor_rechtsgelehrsamkeit02_1758_slower_processors_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_estor_rechtsgelehrsamkeit02_1758_slower_processors_ocr",
+ "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/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_estor_rechtsgelehrsamkeit02_1758_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_estor_rechtsgelehrsamkeit02_1758_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_estor_rechtsgelehrsamkeit02_1758_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_estor_rechtsgelehrsamkeit02_1758_slower_processors_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 63.120688,
+ "cpu_time": 99.46367699999999,
+ "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": 3.802239924888018
+ },
+ "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-data18_frak_complex_bernd_lebensbeschreibung_1738_selected_pages_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_bernd_lebensbeschreibung_1738_selected_pages_ocr",
+ "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/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_bernd_lebensbeschreibung_1738_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_bernd_lebensbeschreibung_1738_selected_pages_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_bernd_lebensbeschreibung_1738_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_bernd_lebensbeschreibung_1738_selected_pages_ocr"
+ },
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 3,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 81.051665,
+ "cpu_time": 123.14032399999999,
+ "cer_mean": 0.08296660948260859,
+ "cer_median": 0.03058103975535168,
+ "cer_range": [
+ 0.02522935779816514,
+ 0.19308943089430894
+ ],
+ "cer_standard_deviation": 0.09540669248903694,
+ "wer": 0.16153231203986018,
+ "pages_per_minute": 2.220805704608289
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.19308943089430894,
+ "wer": 0.38235294117647056
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.03058103975535168,
+ "wer": 0.05357142857142857
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.02522935779816514,
+ "wer": 0.048672566371681415
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data16_ant_complex_alberti_pictura_1540_selected_pages_ocr-eval",
+ "label": "Workflow on data 16_ant_complex_alberti_pictura_1540_selected_pages_ocr",
+ "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/16_ant_complex.ocrd.zip",
+ "label": "GT workspace 16th century Antiqua complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_alberti_pictura_1540_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 16_ant_complex_alberti_pictura_1540_selected_pages_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_alberti_pictura_1540_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_ant_complex_alberti_pictura_1540_selected_pages_ocr"
+ },
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 3,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 69.338131,
+ "cpu_time": 122.13893500000002,
+ "cer_mean": 0.11095656062741122,
+ "cer_median": 0.12244897959183673,
+ "cer_range": [
+ 0.07700101317122594,
+ 0.133419689119171
+ ],
+ "cer_standard_deviation": 0.02991360090611332,
+ "wer": 0.26341315349276156,
+ "pages_per_minute": 2.595974212226747
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0007",
+ "cer": 0.133419689119171,
+ "wer": 0.36363636363636365
+ },
+ {
+ "page_id": "phys_0008",
+ "cer": 0.07700101317122594,
+ "wer": 0.18012422360248448
+ },
+ {
+ "page_id": "phys_0009",
+ "cer": 0.12244897959183673,
+ "wer": 0.24647887323943662
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data18_frak_complex_buerger_gedichte_1778_selected_pages_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_buerger_gedichte_1778_selected_pages_ocr",
+ "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/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_buerger_gedichte_1778_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_buerger_gedichte_1778_selected_pages_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_buerger_gedichte_1778_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_buerger_gedichte_1778_selected_pages_ocr"
+ },
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 2,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 46.097406,
+ "cpu_time": 67.55636299999999,
+ "cer_mean": 0.15032645549695894,
+ "cer_median": 0.15032645549695894,
+ "cer_range": [
+ 0.04932735426008968,
+ 0.2513255567338282
+ ],
+ "cer_standard_deviation": 0.14283429875667375,
+ "wer": 0.3456967605903776,
+ "pages_per_minute": 2.60318335482912
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.2513255567338282,
+ "wer": 0.475177304964539
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.04932735426008968,
+ "wer": 0.21621621621621623
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data16_frak_simple_trota_mordtbrenner_1540_minimal_ocr-eval",
+ "label": "Workflow on data 16_frak_simple_trota_mordtbrenner_1540_minimal_ocr",
+ "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/16_frak_simple.ocrd.zip",
+ "label": "GT workspace 16th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_trota_mordtbrenner_1540_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 16_frak_simple_trota_mordtbrenner_1540_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_trota_mordtbrenner_1540_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_frak_simple_trota_mordtbrenner_1540_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 7.257076,
+ "cpu_time": 8.706393,
+ "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": 16.53558540657422
+ },
+ "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-data18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr-eval",
+ "label": "Workflow on data 18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr",
+ "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/18_ant_simple.ocrd.zip",
+ "label": "GT workspace 18th century Antiqua simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr"
+ },
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 189.99469000000002,
+ "cpu_time": 255.53890399999997,
+ "cer_mean": 0.03947936370768366,
+ "cer_median": 0.0376249265138154,
+ "cer_range": [
+ 0.022336769759450172,
+ 0.05847639484978541
+ ],
+ "cer_standard_deviation": 0.01814103989293333,
+ "wer": 0.15140780173936938,
+ "pages_per_minute": 0.947394898246893
+ },
+ "by_page": [
+ {
+ "page_id": "phys_00003",
+ "cer": 0.022336769759450172,
+ "wer": 0.10970464135021098
+ },
+ {
+ "page_id": "phys_00005",
+ "cer": 0.0376249265138154,
+ "wer": 0.16942148760330578
+ },
+ {
+ "page_id": "phys_00010",
+ "cer": 0.05847639484978541,
+ "wer": 0.17509727626459143
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data17_frak_simple_calvi_beutelschneider01_1627_selected_pages_ocr-eval",
+ "label": "Workflow on data 17_frak_simple_calvi_beutelschneider01_1627_selected_pages_ocr",
+ "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/17_frak_simple.ocrd.zip",
+ "label": "GT workspace 17th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_calvi_beutelschneider01_1627_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 17_frak_simple_calvi_beutelschneider01_1627_selected_pages_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_calvi_beutelschneider01_1627_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 17_frak_simple_calvi_beutelschneider01_1627_selected_pages_ocr"
+ },
+ "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": {
+ "level-of-operation": "page",
+ "dpi": 0,
+ "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 2.6.2)",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Black Letter"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 88.38021099999999,
+ "cpu_time": 127.13157100000001,
+ "cer_mean": 0.047703401203510305,
+ "cer_median": 0.04723502304147465,
+ "cer_range": [
+ 0.04592720970537262,
+ 0.04994797086368366
+ ],
+ "cer_standard_deviation": 0.002050893378518156,
+ "wer": 0.17277993391305632,
+ "pages_per_minute": 2.0366550154536296
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.04994797086368366,
+ "wer": 0.21019108280254778
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.04723502304147465,
+ "wer": 0.15172413793103448
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.04592720970537262,
+ "wer": 0.1564245810055866
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data18_frak_complex_buerger_gedichte_1778_minimal_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_buerger_gedichte_1778_minimal_ocr",
+ "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/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_buerger_gedichte_1778_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_buerger_gedichte_1778_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_buerger_gedichte_1778_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_buerger_gedichte_1778_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 2,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 6.016163,
+ "cpu_time": 7.291227,
+ "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": 19.94626807817541
+ },
+ "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-data18_ant_simple_ballenstedt_delatio_1777_minimal_ocr-eval",
+ "label": "Workflow on data 18_ant_simple_ballenstedt_delatio_1777_minimal_ocr",
+ "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/18_ant_simple.ocrd.zip",
+ "label": "GT workspace 18th century Antiqua simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_ant_simple_ballenstedt_delatio_1777_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 18_ant_simple_ballenstedt_delatio_1777_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_ant_simple_ballenstedt_delatio_1777_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_ant_simple_ballenstedt_delatio_1777_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 24.13474,
+ "cpu_time": 27.011507,
+ "cer_mean": 0.06438361522903834,
+ "cer_median": 0.03969957081545064,
+ "cer_range": [
+ 0.021764032073310423,
+ 0.13168724279835392
+ ],
+ "cer_standard_deviation": 0.058972490200809365,
+ "wer": 0.16906902212925057,
+ "pages_per_minute": 7.458128821773095
+ },
+ "by_page": [
+ {
+ "page_id": "phys_00003",
+ "cer": 0.021764032073310423,
+ "wer": 0.12236286919831224
+ },
+ {
+ "page_id": "phys_00005",
+ "cer": 0.13168724279835392,
+ "wer": 0.2603305785123967
+ },
+ {
+ "page_id": "phys_00010",
+ "cer": 0.03969957081545064,
+ "wer": 0.1245136186770428
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data17_frak_complex_huebner_handbuch_1696_minimal_ocr-eval",
+ "label": "Workflow on data 17_frak_complex_huebner_handbuch_1696_minimal_ocr",
+ "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/17_frak_complex.ocrd.zip",
+ "label": "GT workspace 17th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 17_frak_complex_huebner_handbuch_1696_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 17_frak_complex_huebner_handbuch_1696_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 3,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 16.954402,
+ "cpu_time": 19.205537,
+ "cer_mean": 0.20627969569747484,
+ "cer_median": 0.08702290076335878,
+ "cer_range": [
+ 0.08487084870848709,
+ 0.44694533762057875
+ ],
+ "cer_standard_deviation": 0.20842533731220508,
+ "wer": 0.3271033769383102,
+ "pages_per_minute": 10.616711813250621
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.44694533762057875,
+ "wer": 0.6341463414634146
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.08702290076335878,
+ "wer": 0.1504424778761062
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.08487084870848709,
+ "wer": 0.19672131147540983
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data19_frak_simple_arnimb_goethe03_1835_slower_processors_ocr-eval",
+ "label": "Workflow on data 19_frak_simple_arnimb_goethe03_1835_slower_processors_ocr",
+ "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/19_frak_simple.ocrd.zip",
+ "label": "GT workspace 19th century Black letter simple layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_arnimb_goethe03_1835_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 19_frak_simple_arnimb_goethe03_1835_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_arnimb_goethe03_1835_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 19_frak_simple_arnimb_goethe03_1835_slower_processors_ocr"
+ },
+ "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": [
+ "Black Letter"
+ ],
+ "publication_century": "1800-1900",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 1,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 12.164379999999998,
+ "cpu_time": 19.670232,
+ "cer_mean": 0.0056657223796034,
+ "cer_median": 0.0056657223796034,
+ "cer_range": [
+ 0.0056657223796034,
+ 0.0056657223796034
+ ],
+ "cer_standard_deviation": null,
+ "wer": 0.031746031746031744,
+ "pages_per_minute": 4.932433876613524
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.0056657223796034,
+ "wer": 0.031746031746031744
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data17_frak_complex_silesius_seelenlust01_1657_slower_processors_ocr-eval",
+ "label": "Workflow on data 17_frak_complex_silesius_seelenlust01_1657_slower_processors_ocr",
+ "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/17_frak_complex.ocrd.zip",
+ "label": "GT workspace 17th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 17_frak_complex_silesius_seelenlust01_1657_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 17_frak_complex_silesius_seelenlust01_1657_slower_processors_ocr"
+ },
+ "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,
@@ -1734,67 +6949,63 @@
"document_metadata": {
"data_properties": {
"fonts": [
+ "Antiqua",
"Black Letter"
],
- "publication_century": "1500-1600",
+ "publication_century": "1600-1700",
"publication_decade": "",
- "publication_year": "16th century",
- "number_of_pages": 6,
- "layout": "simple"
+ "publication_year": "17th century",
+ "number_of_pages": 5,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 24.570796,
- "cpu_time": 30.071177,
- "cer_mean": 0.13490219630962338,
- "cer_median": 0.08500485504230823,
+ "wall_time": 51.051955,
+ "cpu_time": 79.096845,
+ "cer_mean": 0.16456973001701142,
+ "cer_median": 0.11229135053110774,
"cer_range": [
- 0.014285714285714285,
- 0.4184652278177458
+ 0.05917159763313609,
+ 0.34146341463414637
],
- "cer_standard_deviation": 0.14421604715728026,
- "wer": 0.2783759515546502,
- "pages_per_minute": 14.651539982668854
+ "cer_standard_deviation": 0.11554307430913685,
+ "wer": 0.34312458192490203,
+ "pages_per_minute": 5.876366536795701
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.0718562874251497,
- "wer": 0.19469026548672566
+ "cer": 0.05917159763313609,
+ "wer": 0.2463768115942029
},
{
"page_id": "phys_0002",
- "cer": 0.014285714285714285,
- "wer": 0.0196078431372549
+ "cer": 0.0913884007029877,
+ "wer": 0.28125
},
{
"page_id": "phys_0003",
- "cer": 0.4184652278177458,
- "wer": 0.48333333333333334
- },
- {
- "page_id": "phys_0007",
- "cer": 0.09135802469135802,
- "wer": 0.33495145631067963
+ "cer": 0.34146341463414637,
+ "wer": 0.5774647887323944
},
{
- "page_id": "phys_0021",
- "cer": 0.13479623824451412,
- "wer": 0.375
+ "page_id": "phys_0004",
+ "cer": 0.11229135053110774,
+ "wer": 0.1693548387096774
},
{
- "page_id": "phys_0029",
- "cer": 0.07865168539325842,
- "wer": 0.2626728110599078
+ "page_id": "phys_0005",
+ "cer": 0.21853388658367912,
+ "wer": 0.4411764705882353
}
]
}
},
{
- "eval_workflow_id": "wf-data18_fontmix_complex_minimal_ocr-eval",
- "label": "Workflow on data 18_fontmix_complex_minimal_ocr",
+ "eval_workflow_id": "wf-data17_frak_complex_silesius_seelenlust01_1657_minimal_ocr-eval",
+ "label": "Workflow on data 17_frak_complex_silesius_seelenlust01_1657_minimal_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -1805,16 +7016,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_fontmix_complex.ocrd.zip",
- "label": "GT workspace 18th century Font Mix complex layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/17_frak_complex.ocrd.zip",
+ "label": "GT workspace 17th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 18_fontmix_complex_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 17_frak_complex_silesius_seelenlust01_1657_minimal_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_fontmix_complex_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 17_frak_complex_silesius_seelenlust01_1657_minimal_ocr"
},
"workflow_steps": [
{
@@ -1852,90 +7063,175 @@
"Antiqua",
"Black Letter"
],
- "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": 5,
"layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 18.971287,
- "cpu_time": 22.54346,
- "cer_mean": 0.20433470587543373,
- "cer_median": 0.1700173533179325,
+ "wall_time": 13.146098,
+ "cpu_time": 16.982357,
+ "cer_mean": 0.3024629978548877,
+ "cer_median": 0.2951219512195122,
"cer_range": [
- 0.04063745019920319,
- 0.43666666666666665
+ 0.19271623672230653,
+ 0.44970414201183434
],
- "cer_standard_deviation": 0.19097881579976753,
- "wer": 0.31449770246107817,
- "pages_per_minute": 12.6506968135583
+ "cer_standard_deviation": 0.10474043149462715,
+ "wer": 0.5195139299620842,
+ "pages_per_minute": 22.820459728810782
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.43666666666666665,
- "wer": 0.583941605839416
+ "cer": 0.44970414201183434,
+ "wer": 0.7101449275362319
},
{
"page_id": "phys_0002",
- "cer": 0.04063745019920319,
- "wer": 0.10344827586206896
+ "cer": 0.2179261862917399,
+ "wer": 0.5208333333333334
},
{
"page_id": "phys_0003",
- "cer": 0.05536912751677853,
- "wer": 0.18435754189944134
+ "cer": 0.2951219512195122,
+ "wer": 0.4647887323943662
},
{
"page_id": "phys_0004",
- "cer": 0.28466557911908646,
- "wer": 0.3862433862433862
+ "cer": 0.19271623672230653,
+ "wer": 0.3870967741935484
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 0.35684647302904565,
+ "wer": 0.5147058823529411
}
]
}
},
{
- "eval_workflow_id": "wf-data16_ant_simple_minimal_ocr-eval",
- "label": "Workflow on data 16_ant_simple_minimal_ocr",
+ "eval_workflow_id": "wf-data18_frak_complex_luz_blitz_1784_slower_processors_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_luz_blitz_1784_slower_processors_ocr",
"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/16_ant_simple.ocrd.zip",
- "label": "GT workspace 16th century Antiqua simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 16_ant_simple_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_luz_blitz_1784_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_luz_blitz_1784_slower_processors_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_ant_simple_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_luz_blitz_1784_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_luz_blitz_1784_slower_processors_ocr"
},
"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,
@@ -1955,52 +7251,58 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua"
+ "Antiqua",
+ "Black Letter"
],
- "publication_century": "1500-1600",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "16th century",
- "number_of_pages": 3,
- "layout": "simple"
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 7.865648,
- "cpu_time": 10.22416,
- "cer_mean": 0.07452119312897007,
- "cer_median": 0.0629800307219662,
+ "wall_time": 64.58104600000001,
+ "cpu_time": 99.479389,
+ "cer_mean": 0.02547899799369996,
+ "cer_median": 0.027204160076875337,
"cer_range": [
- 0.037037037037037035,
- 0.12354651162790697
+ 0.010332950631458095,
+ 0.03717472118959108
],
- "cer_standard_deviation": 0.044394494261965886,
- "wer": 0.22683890077340793,
- "pages_per_minute": 22.88431925761234
+ "cer_standard_deviation": 0.011266997500136374,
+ "wer": 0.07714749104131867,
+ "pages_per_minute": 3.716260650222357
},
"by_page": [
{
- "page_id": "phys_0007",
- "cer": 0.0629800307219662,
- "wer": 0.21359223300970873
+ "page_id": "phys_0001",
+ "cer": 0.03717472118959108,
+ "wer": 0.10256410256410256
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.025119617224880382,
+ "wer": 0.09565217391304348
},
{
- "page_id": "phys_0013",
- "cer": 0.12354651162790697,
- "wer": 0.33980582524271846
+ "page_id": "phys_0003",
+ "cer": 0.029288702928870293,
+ "wer": 0.07924528301886792
},
{
- "page_id": "phys_0014",
- "cer": 0.037037037037037035,
- "wer": 0.1271186440677966
+ "page_id": "phys_0004",
+ "cer": 0.010332950631458095,
+ "wer": 0.0311284046692607
}
]
}
},
{
- "eval_workflow_id": "wf-data19_ant_simple_selected_pages_ocr-eval",
- "label": "Workflow on data 19_ant_simple_selected_pages_ocr",
+ "eval_workflow_id": "wf-data19_ant_simple_blumenbach_anatomie_1805_selected_pages_ocr-eval",
+ "label": "Workflow on data 19_ant_simple_blumenbach_anatomie_1805_selected_pages_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
@@ -2015,12 +7317,12 @@
"label": "GT workspace 19th century Antiqua simple layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 19_ant_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_blumenbach_anatomie_1805_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 19_ant_simple_blumenbach_anatomie_1805_selected_pages_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 19_ant_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_blumenbach_anatomie_1805_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 19_ant_simple_blumenbach_anatomie_1805_selected_pages_ocr"
},
"workflow_steps": [
{
@@ -2123,7 +7425,8 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua"
+ "Antiqua",
+ "Black Letter"
],
"publication_century": "1800-1900",
"publication_decade": "",
@@ -2135,8 +7438,8 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 113.83170799999999,
- "cpu_time": 159.490409,
+ "wall_time": 97.867795,
+ "cpu_time": 142.045967,
"cer_mean": 0.08457767160660273,
"cer_median": 0.07737397420867527,
"cer_range": [
@@ -2145,7 +7448,7 @@
],
"cer_standard_deviation": 0.046973440956685215,
"wer": 0.3081861348528015,
- "pages_per_minute": 1.5812817286375076
+ "pages_per_minute": 1.8392158523649174
},
"by_page": [
{
@@ -2167,8 +7470,8 @@
}
},
{
- "eval_workflow_id": "wf-data19_ant_simple_slower_processors_ocr-eval",
- "label": "Workflow on data 19_ant_simple_slower_processors_ocr",
+ "eval_workflow_id": "wf-data18_frak_complex_nn_besuch_1780_slower_processors_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_nn_besuch_1780_slower_processors_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
@@ -2179,16 +7482,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/19_ant_simple.ocrd.zip",
- "label": "GT workspace 19th century Antiqua simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 19_ant_simple_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_nn_besuch_1780_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_nn_besuch_1780_slower_processors_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 19_ant_simple_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_nn_besuch_1780_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_nn_besuch_1780_slower_processors_ocr"
},
"workflow_steps": [
{
@@ -2303,52 +7606,345 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua"
+ "Antiqua",
+ "Black Letter"
],
- "publication_century": "1800-1900",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "19th century",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 49.367179,
+ "cpu_time": 71.436347,
+ "cer_mean": 0.07900170040044713,
+ "cer_median": 0.018462174141687326,
+ "cer_range": [
+ 0.013259668508287293,
+ 0.26582278481012656
+ ],
+ "cer_standard_deviation": 0.1245848990441196,
+ "wer": 0.13667696498571524,
+ "pages_per_minute": 4.8615295599531825
+ },
+ "by_page": [
+ {
+ "page_id": "phys_00001",
+ "cer": 0.26582278481012656,
+ "wer": 0.3620689655172414
+ },
+ {
+ "page_id": "phys_00002",
+ "cer": 0.013259668508287293,
+ "wer": 0.057803468208092484
+ },
+ {
+ "page_id": "phys_00003",
+ "cer": 0.020697167755991286,
+ "wer": 0.06936416184971098
+ },
+ {
+ "page_id": "phys_00004",
+ "cer": 0.016227180527383367,
+ "wer": 0.05747126436781609
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-data17_frak_complex_huebner_handbuch_1696_slower_processors_ocr-eval",
+ "label": "Workflow on data 17_frak_complex_huebner_handbuch_1696_slower_processors_ocr",
+ "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/17_frak_complex.ocrd.zip",
+ "label": "GT workspace 17th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 17_frak_complex_huebner_handbuch_1696_slower_processors_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 17_frak_complex_huebner_handbuch_1696_slower_processors_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
"number_of_pages": 3,
- "layout": "simple"
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 61.992684,
+ "cpu_time": 83.585045,
+ "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": 2.903568427526061
+ },
+ "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-data18_frak_complex_justi_abhandlung01_1758_minimal_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_justi_abhandlung01_1758_minimal_ocr",
+ "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/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_justi_abhandlung01_1758_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_justi_abhandlung01_1758_minimal_ocr"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_justi_abhandlung01_1758_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_justi_abhandlung01_1758_minimal_ocr"
+ },
+ "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",
+ "Black Letter"
+ ],
+ "publication_century": "1700-1800",
+ "publication_decade": "",
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 47.942433,
- "cpu_time": 73.79890900000001,
- "cer_mean": 0.08697690719872657,
- "cer_median": 0.04421052631578947,
+ "wall_time": 14.25996,
+ "cpu_time": 17.707908,
+ "cer_mean": 0.1347518651788604,
+ "cer_median": 0.14344023364180397,
"cer_range": [
- 0.03735325506937033,
- 0.17936694021101993
+ 0.09893550407013149,
+ 0.15319148936170213
],
- "cer_standard_deviation": 0.08008554296654442,
- "wer": 0.2764482431149098,
- "pages_per_minute": 3.7545028221659087
+ "cer_standard_deviation": 0.02557054025937362,
+ "wer": 0.19715777371165608,
+ "pages_per_minute": 16.830341740089032
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.04421052631578947,
- "wer": 0.16666666666666666
+ "cer": 0.13375130616509928,
+ "wer": 0.2
},
{
"page_id": "phys_0002",
- "cer": 0.03735325506937033,
- "wer": 0.17037037037037037
+ "cer": 0.15319148936170213,
+ "wer": 0.22916666666666666
},
{
"page_id": "phys_0003",
- "cer": 0.17936694021101993,
- "wer": 0.49230769230769234
+ "cer": 0.09893550407013149,
+ "wer": 0.12598425196850394
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.15312916111850866,
+ "wer": 0.23348017621145375
}
]
}
},
{
- "eval_workflow_id": "wf-data19_frak_simple_selected_pages_ocr-eval",
- "label": "Workflow on data 19_frak_simple_selected_pages_ocr",
+ "eval_workflow_id": "wf-data18_frak_complex_euler_rechenkunst01_1738_selected_pages_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_euler_rechenkunst01_1738_selected_pages_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
@@ -2359,16 +7955,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/19_frak_simple.ocrd.zip",
- "label": "GT workspace 19th century Black letter simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 19_frak_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_euler_rechenkunst01_1738_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_euler_rechenkunst01_1738_selected_pages_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 19_frak_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_euler_rechenkunst01_1738_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_euler_rechenkunst01_1738_selected_pages_ocr"
},
"workflow_steps": [
{
@@ -2407,7 +8003,7 @@
"method": "li",
"level-of-operation": "page",
"dpi": 0,
- "window_size": 301,
+ "window_size": 201,
"k": 0.34
}
},
@@ -2471,42 +8067,68 @@
"document_metadata": {
"data_properties": {
"fonts": [
+ "Antiqua",
"Black Letter"
],
- "publication_century": "1800-1900",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "19th century",
- "number_of_pages": 1,
- "layout": "simple"
+ "publication_year": "18th century",
+ "number_of_pages": 6,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 50.242278999999996,
- "cpu_time": 72.25566799999999,
- "cer_mean": 0.007554296506137866,
- "cer_median": 0.007554296506137866,
+ "wall_time": 161.89715999999999,
+ "cpu_time": 264.56102,
+ "cer_mean": 0.1953749490770702,
+ "cer_median": 0.1576184687985986,
"cer_range": [
- 0.007554296506137866,
- 0.007554296506137866
+ 0.09627059843885516,
+ 0.3873684210526316
],
- "cer_standard_deviation": null,
- "wer": 0.015873015873015872,
- "pages_per_minute": 1.1942133437060052
+ "cer_standard_deviation": 0.11209255429867662,
+ "wer": 0.3314757763924516,
+ "pages_per_minute": 2.2236338179125563
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.007554296506137866,
- "wer": 0.015873015873015872
+ "cer": 0.09627059843885516,
+ "wer": 0.13953488372093023
+ },
+ {
+ "page_id": "phys_0002",
+ "cer": 0.3873684210526316,
+ "wer": 0.6036585365853658
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.12221368178324366,
+ "wer": 0.25728155339805825
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.2595959595959596,
+ "wer": 0.36129032258064514
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 0.11377777777777778,
+ "wer": 0.28402366863905326
+ },
+ {
+ "page_id": "phys_0006",
+ "cer": 0.1930232558139535,
+ "wer": 0.34306569343065696
}
]
}
},
{
- "eval_workflow_id": "wf-data17_frak_simple_selected_pages_ocr-eval",
- "label": "Workflow on data 17_frak_simple_selected_pages_ocr",
+ "eval_workflow_id": "wf-data16_ant_simple_heyden_paedono_1548_selected_pages_ocr-eval",
+ "label": "Workflow on data 16_ant_simple_heyden_paedono_1548_selected_pages_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
@@ -2517,16 +8139,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/17_frak_simple.ocrd.zip",
- "label": "GT workspace 17th century Black letter simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/16_ant_simple.ocrd.zip",
+ "label": "GT workspace 16th century Antiqua simple layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 17_frak_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_heyden_paedono_1548_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 16_ant_simple_heyden_paedono_1548_selected_pages_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 17_frak_simple_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_heyden_paedono_1548_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 16_ant_simple_heyden_paedono_1548_selected_pages_ocr"
},
"workflow_steps": [
{
@@ -2565,7 +8187,7 @@
"method": "li",
"level-of-operation": "page",
"dpi": 0,
- "window_size": 133,
+ "window_size": 301,
"k": 0.34
}
},
@@ -2629,172 +8251,73 @@
"document_metadata": {
"data_properties": {
"fonts": [
+ "Antiqua",
"Black Letter"
],
- "publication_century": "1600-1700",
- "publication_decade": "",
- "publication_year": "17th century",
- "number_of_pages": 3,
- "layout": "simple"
- }
- }
- },
- "evaluation_results": {
- "document_wide": {
- "wall_time": 74.155102,
- "cpu_time": 112.343518,
- "cer_mean": 0.047703401203510305,
- "cer_median": 0.04723502304147465,
- "cer_range": [
- 0.04592720970537262,
- 0.04994797086368366
- ],
- "cer_standard_deviation": 0.002050893378518156,
- "wer": 0.17277993391305632,
- "pages_per_minute": 2.427344783370401
- },
- "by_page": [
- {
- "page_id": "phys_0001",
- "cer": 0.04994797086368366,
- "wer": 0.21019108280254778
- },
- {
- "page_id": "phys_0002",
- "cer": 0.04723502304147465,
- "wer": 0.15172413793103448
- },
- {
- "page_id": "phys_0003",
- "cer": 0.04592720970537262,
- "wer": 0.1564245810055866
- }
- ]
- }
- },
- {
- "eval_workflow_id": "wf-data16_ant_complex_minimal_ocr-eval",
- "label": "Workflow on data 16_ant_complex_minimal_ocr",
- "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/16_ant_complex.ocrd.zip",
- "label": "GT workspace 16th century Antiqua complex layout"
- },
- "ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 16_ant_complex_minimal_ocr"
- },
- "eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_ant_complex_minimal_ocr"
- },
- "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": "1500-1600",
"publication_decade": "",
"publication_year": "16th century",
"number_of_pages": 3,
- "layout": "complex"
+ "layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 7.579529,
- "cpu_time": 10.163754,
- "cer_mean": 0.10240852523716282,
- "cer_median": 0.10536980749746708,
+ "wall_time": 72.48043099999998,
+ "cpu_time": 107.649963,
+ "cer_mean": 0.05488709037929677,
+ "cer_median": 0.03488372093023256,
"cer_range": [
- 0.07124352331606218,
- 0.1306122448979592
+ 0.028395061728395062,
+ 0.10138248847926268
],
- "cer_standard_deviation": 0.02979493530847308,
- "wer": 0.23466068901129858,
- "pages_per_minute": 23.74817749229537
+ "cer_standard_deviation": 0.04039668560556835,
+ "wer": 0.13745817563490756,
+ "pages_per_minute": 2.483428940978566
},
"by_page": [
{
"page_id": "phys_0007",
- "cer": 0.07124352331606218,
- "wer": 0.2231404958677686
+ "cer": 0.10138248847926268,
+ "wer": 0.17475728155339806
},
{
- "page_id": "phys_0008",
- "cer": 0.10536980749746708,
- "wer": 0.2484472049689441
+ "page_id": "phys_0013",
+ "cer": 0.03488372093023256,
+ "wer": 0.13592233009708737
},
{
- "page_id": "phys_0009",
- "cer": 0.1306122448979592,
- "wer": 0.2323943661971831
+ "page_id": "phys_0014",
+ "cer": 0.028395061728395062,
+ "wer": 0.1016949152542373
}
]
}
},
{
- "eval_workflow_id": "wf-data17_frak_simple_slower_processors_ocr-eval",
- "label": "Workflow on data 17_frak_simple_slower_processors_ocr",
+ "eval_workflow_id": "wf-data17_frak_complex_huebner_handbuch_1696_selected_pages_ocr-eval",
+ "label": "Workflow on data 17_frak_complex_huebner_handbuch_1696_selected_pages_ocr",
"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"
+ "@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/17_frak_simple.ocrd.zip",
- "label": "GT workspace 17th century Black letter simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/17_frak_complex.ocrd.zip",
+ "label": "GT workspace 17th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 17_frak_simple_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 17_frak_complex_huebner_handbuch_1696_selected_pages_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 17_frak_simple_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 17_frak_complex_huebner_handbuch_1696_selected_pages_ocr"
},
"workflow_steps": [
{
@@ -2828,133 +8351,122 @@
}
},
{
- "id": "ocrd-cis-ocropy-denoise",
+ "id": "ocrd-skimage-binarize",
"params": {
+ "method": "li",
"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
+ "window_size": 301,
+ "k": 0.34
}
},
{
- "id": "ocrd-tesserocr-segment",
+ "id": "ocrd-skimage-denoise",
"params": {
+ "level-of-operation": "page",
"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"
+ "protect": 0.0,
+ "maxsize": 1.0
}
},
{
- "id": "ocrd-cis-ocropy-dewarp",
+ "id": "ocrd-tesserocr-deskew",
"params": {
+ "operation_level": "page",
"dpi": 0,
- "range": 4.0,
- "smoothness": 1.0,
- "max_neighbour": 0.05
+ "min_orientation_confidence": 1.5
}
},
{
- "id": "ocrd-tesserocr-recognize",
+ "id": "ocrd-cis-ocropy-segment",
"params": {
- "textequiv_level": "glyph",
- "overwrite_segments": true,
- "model": "Fraktur_GT4HistOCR",
+ "level-of-operation": "page",
"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"
+ "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 2.6.2)",
"document_metadata": {
"data_properties": {
"fonts": [
+ "Antiqua",
"Black Letter"
],
"publication_century": "1600-1700",
"publication_decade": "",
"publication_year": "17th century",
"number_of_pages": 3,
- "layout": "simple"
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 45.753605,
- "cpu_time": 66.773523,
- "cer_mean": 0.15501219595245352,
- "cer_median": 0.11654526534859522,
+ "wall_time": 182.12772,
+ "cpu_time": 235.173884,
+ "cer_mean": 0.2294897403822962,
+ "cer_median": 0.06717557251908397,
"cer_range": [
- 0.09618717504332755,
- 0.2523041474654378
+ 0.061808118081180814,
+ 0.5594855305466238
],
- "cer_standard_deviation": 0.08486993479509113,
- "wer": 0.360948415946103,
- "pages_per_minute": 3.9341162297484535
+ "cer_standard_deviation": 0.2857973382248686,
+ "wer": 0.36395117469236493,
+ "pages_per_minute": 0.9883174291096379
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.11654526534859522,
- "wer": 0.46496815286624205
+ "cer": 0.5594855305466238,
+ "wer": 0.8292682926829268
},
{
"page_id": "phys_0002",
- "cer": 0.2523041474654378,
- "wer": 0.4
+ "cer": 0.06717557251908397,
+ "wer": 0.11504424778761062
},
{
"page_id": "phys_0003",
- "cer": 0.09618717504332755,
- "wer": 0.21787709497206703
+ "cer": 0.061808118081180814,
+ "wer": 0.14754098360655737
}
]
}
},
{
- "eval_workflow_id": "wf-data16_ant_complex_slower_processors_ocr-eval",
- "label": "Workflow on data 16_ant_complex_slower_processors_ocr",
+ "eval_workflow_id": "wf-data17_frak_simple_calvi_beutelschneider01_1627_slower_processors_ocr-eval",
+ "label": "Workflow on data 17_frak_simple_calvi_beutelschneider01_1627_slower_processors_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
@@ -2965,16 +8477,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/16_ant_complex.ocrd.zip",
- "label": "GT workspace 16th century Antiqua complex layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/17_frak_simple.ocrd.zip",
+ "label": "GT workspace 17th century Black letter simple layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 16_ant_complex_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_calvi_beutelschneider01_1627_slower_processors_ocr_ocr.zip",
+ "label": "OCR workspace for 17_frak_simple_calvi_beutelschneider01_1627_slower_processors_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_ant_complex_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_calvi_beutelschneider01_1627_slower_processors_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 17_frak_simple_calvi_beutelschneider01_1627_slower_processors_ocr"
},
"workflow_steps": [
{
@@ -3089,52 +8601,53 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua"
+ "Antiqua",
+ "Black Letter"
],
- "publication_century": "1500-1600",
+ "publication_century": "1600-1700",
"publication_decade": "",
- "publication_year": "16th century",
+ "publication_year": "17th century",
"number_of_pages": 3,
- "layout": "complex"
+ "layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 32.228635,
- "cpu_time": 53.661828,
- "cer_mean": 0.10363204260868718,
- "cer_median": 0.11836734693877551,
+ "wall_time": 57.283714,
+ "cpu_time": 83.17899299999999,
+ "cer_mean": 0.15501219595245352,
+ "cer_median": 0.11654526534859522,
"cer_range": [
- 0.046632124352331605,
- 0.1458966565349544
+ 0.09618717504332755,
+ 0.2523041474654378
],
- "cer_standard_deviation": 0.05124654849483992,
- "wer": 0.2145458690579216,
- "pages_per_minute": 5.585095366279088
+ "cer_standard_deviation": 0.08486993479509113,
+ "wer": 0.360948415946103,
+ "pages_per_minute": 3.142254358717034
},
"by_page": [
{
- "page_id": "phys_0007",
- "cer": 0.046632124352331605,
- "wer": 0.1652892561983471
+ "page_id": "phys_0001",
+ "cer": 0.11654526534859522,
+ "wer": 0.46496815286624205
},
{
- "page_id": "phys_0008",
- "cer": 0.1458966565349544,
- "wer": 0.2670807453416149
+ "page_id": "phys_0002",
+ "cer": 0.2523041474654378,
+ "wer": 0.4
},
{
- "page_id": "phys_0009",
- "cer": 0.11836734693877551,
- "wer": 0.2112676056338028
+ "page_id": "phys_0003",
+ "cer": 0.09618717504332755,
+ "wer": 0.21787709497206703
}
]
}
},
{
- "eval_workflow_id": "wf-data17_frak_simple_minimal_ocr-eval",
- "label": "Workflow on data 17_frak_simple_minimal_ocr",
+ "eval_workflow_id": "wf-data18_frak_complex_estor_rechtsgelehrsamkeit02_1758_minimal_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_estor_rechtsgelehrsamkeit02_1758_minimal_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -3145,16 +8658,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/17_frak_simple.ocrd.zip",
- "label": "GT workspace 17th century Black letter simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 17_frak_simple_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_estor_rechtsgelehrsamkeit02_1758_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_estor_rechtsgelehrsamkeit02_1758_minimal_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 17_frak_simple_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_estor_rechtsgelehrsamkeit02_1758_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_estor_rechtsgelehrsamkeit02_1758_minimal_ocr"
},
"workflow_steps": [
{
@@ -3189,52 +8702,58 @@
"document_metadata": {
"data_properties": {
"fonts": [
+ "Antiqua",
"Black Letter"
],
- "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": 7.538366,
- "cpu_time": 10.064678,
- "cer_mean": 0.08427970798416445,
- "cer_median": 0.09331797235023041,
+ "wall_time": 18.505333,
+ "cpu_time": 22.105623,
+ "cer_mean": 0.11589408928020027,
+ "cer_median": 0.10084584323499293,
"cer_range": [
- 0.05025996533795494,
- 0.10926118626430802
+ 0.05331088664421998,
+ 0.2085737840065952
],
- "cer_standard_deviation": 0.030521364398782277,
- "wer": 0.1842438386542771,
- "pages_per_minute": 23.87785363565526
+ "cer_standard_deviation": 0.07259145757108061,
+ "wer": 0.20102650242627845,
+ "pages_per_minute": 12.969234328287957
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.10926118626430802,
- "wer": 0.2229299363057325
+ "cer": 0.06310013717421124,
+ "wer": 0.11403508771929824
},
{
"page_id": "phys_0002",
- "cer": 0.09331797235023041,
- "wer": 0.20689655172413793
+ "cer": 0.13859154929577464,
+ "wer": 0.22340425531914893
},
{
"page_id": "phys_0003",
- "cer": 0.05025996533795494,
- "wer": 0.12290502793296089
+ "cer": 0.05331088664421998,
+ "wer": 0.13333333333333333
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.2085737840065952,
+ "wer": 0.3333333333333333
}
]
}
},
{
- "eval_workflow_id": "wf-data16_ant_complex_selected_pages_ocr-eval",
- "label": "Workflow on data 16_ant_complex_selected_pages_ocr",
+ "eval_workflow_id": "wf-data17_frak_complex_silesius_seelenlust01_1657_selected_pages_ocr-eval",
+ "label": "Workflow on data 17_frak_complex_silesius_seelenlust01_1657_selected_pages_ocr",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
@@ -3245,16 +8764,16 @@
"label": "Evaluation Workflow dinglehopper_eval"
},
"gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/16_ant_complex.ocrd.zip",
- "label": "GT workspace 16th century Antiqua complex layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/17_frak_complex.ocrd.zip",
+ "label": "GT workspace 17th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 16_ant_complex_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_selected_pages_ocr_ocr.zip",
+ "label": "OCR workspace for 17_frak_complex_silesius_seelenlust01_1657_selected_pages_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_ant_complex_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_selected_pages_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 17_frak_complex_silesius_seelenlust01_1657_selected_pages_ocr"
},
"workflow_steps": [
{
@@ -3357,167 +8876,98 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua"
+ "Antiqua",
+ "Black Letter"
],
- "publication_century": "1500-1600",
+ "publication_century": "1600-1700",
"publication_decade": "",
- "publication_year": "16th century",
- "number_of_pages": 3,
+ "publication_year": "17th century",
+ "number_of_pages": 5,
"layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 70.23542499999999,
- "cpu_time": 122.07752,
- "cer_mean": 0.11095656062741122,
- "cer_median": 0.12244897959183673,
+ "wall_time": 108.68203399999999,
+ "cpu_time": 154.24126800000005,
+ "cer_mean": 0.2924684311074824,
+ "cer_median": 0.2583479789103691,
"cer_range": [
- 0.07700101317122594,
- 0.133419689119171
+ 0.06676783004552352,
+ 0.7536585365853659
],
- "cer_standard_deviation": 0.02991360090611332,
- "wer": 0.26341315349276156,
- "pages_per_minute": 2.5628092945974204
+ "cer_standard_deviation": 0.2774715020464407,
+ "wer": 0.4982391606465789,
+ "pages_per_minute": 2.7603458360008246
},
"by_page": [
{
- "page_id": "phys_0007",
- "cer": 0.133419689119171,
- "wer": 0.36363636363636365
+ "page_id": "phys_0001",
+ "cer": 0.08481262327416174,
+ "wer": 0.43478260869565216
},
{
- "page_id": "phys_0008",
- "cer": 0.07700101317122594,
- "wer": 0.18012422360248448
+ "page_id": "phys_0002",
+ "cer": 0.2583479789103691,
+ "wer": 0.5625
},
{
- "page_id": "phys_0009",
- "cer": 0.12244897959183673,
- "wer": 0.24647887323943662
+ "page_id": "phys_0003",
+ "cer": 0.7536585365853659,
+ "wer": 0.9014084507042254
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.06676783004552352,
+ "wer": 0.04838709677419355
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 0.2987551867219917,
+ "wer": 0.5441176470588235
}
]
}
},
{
- "eval_workflow_id": "wf-data18_fontmix_complex_slower_processors_ocr-eval",
- "label": "Workflow on data 18_fontmix_complex_slower_processors_ocr",
+ "eval_workflow_id": "wf-data18_frak_complex_euler_rechenkunst01_1738_minimal_ocr-eval",
+ "label": "Workflow on data 18_frak_complex_euler_rechenkunst01_1738_minimal_ocr",
"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"
+ "@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/18_fontmix_complex.ocrd.zip",
- "label": "GT workspace 18th century Font Mix complex layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/18_frak_complex.ocrd.zip",
+ "label": "GT workspace 18th century Black letter complex layout"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 18_fontmix_complex_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_euler_rechenkunst01_1738_minimal_ocr_ocr.zip",
+ "label": "OCR workspace for 18_frak_complex_euler_rechenkunst01_1738_minimal_ocr"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_fontmix_complex_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_fontmix_complex_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_euler_rechenkunst01_1738_minimal_ocr_evaluation.zip",
+ "label": "Evaluation workspace for 18_frak_complex_euler_rechenkunst01_1738_minimal_ocr"
},
"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,
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
"model": "Fraktur_GT4HistOCR",
"dpi": 0,
"padding": 0,
- "segmentation_level": "word",
+ "overwrite_segments": false,
"overwrite_text": true,
"shrink_polygons": false,
"block_polygons": false,
- "find_tables": true,
"find_staves": false,
"sparse_text": false,
"raw_lines": false,
@@ -3543,45 +8993,55 @@
"publication_century": "1700-1800",
"publication_decade": "",
"publication_year": "18th century",
- "number_of_pages": 4,
+ "number_of_pages": 6,
"layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 72.53919699999999,
- "cpu_time": 109.473651,
- "cer_mean": 0.16726583056278752,
- "cer_median": 0.09637318392327315,
+ "wall_time": 18.674437,
+ "cpu_time": 24.291901,
+ "cer_mean": 0.25835849983393794,
+ "cer_median": 0.230420483908856,
"cer_range": [
- 0.03187250996015936,
- 0.4444444444444444
+ 0.08586296617519515,
+ 0.5747368421052632
],
- "cer_standard_deviation": 0.18889822286887584,
- "wer": 0.28912998545359864,
- "pages_per_minute": 3.3085560624554478
+ "cer_standard_deviation": 0.16915619139008603,
+ "wer": 0.37827288927088376,
+ "pages_per_minute": 19.277689603172508
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.4444444444444444,
- "wer": 0.6204379562043796
+ "cer": 0.08586296617519515,
+ "wer": 0.10465116279069768
},
{
"page_id": "phys_0002",
- "cer": 0.03187250996015936,
- "wer": 0.10919540229885058
+ "cer": 0.5747368421052632,
+ "wer": 0.7987804878048781
},
{
"page_id": "phys_0003",
- "cer": 0.1266778523489933,
- "wer": 0.2681564245810056
+ "cer": 0.2767102229054573,
+ "wer": 0.3786407766990291
},
{
"page_id": "phys_0004",
- "cer": 0.06606851549755302,
- "wer": 0.15873015873015872
+ "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
}
]
}
diff --git a/default_data_sources.txt b/default_data_sources.txt
new file mode 100644
index 0000000000000000000000000000000000000000..d9b1ddd6ce17bbc7217efee73c9ee4885ccfb031
--- /dev/null
+++ b/default_data_sources.txt
@@ -0,0 +1,12 @@
+https://github.com/tboenig/16_frak_simple
+https://github.com/tboenig/17_frak_simple
+https://github.com/tboenig/17_frak_complex
+https://github.com/tboenig/18_frak_simple
+https://github.com/tboenig/18_frak_complex
+https://github.com/tboenig/19_frak_simple
+https://github.com/tboenig/16_ant_simple
+https://github.com/tboenig/16_ant_complex
+https://github.com/tboenig/18_ant_simple
+https://github.com/tboenig/19_ant_simple
+https://github.com/tboenig/17_fontmix_simple
+https://github.com/tboenig/18_fontmix_complex
diff --git a/docker-compose.yml b/docker-compose.yml
index 21b37d5d5ef6ec0f2cc52a8b36d680199fc87470..c4be3efcab37ce0cfade9e4bf3683676c53c90fa 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -1,14 +1,16 @@
version: '3.4'
services:
- benchmarks:
+ app:
build:
context: .
dockerfile: Dockerfile
working_dir: /app
volumes:
- ./data:/app/data # this will write the results to your host system
+ - ./gt:/app/gt
# mount your modules, custom workflows and data here
- # - /path/to/models/on/host:/usr/local/share/ocrd-resources/
+ - ./models:/usr/local/share/ocrd-resources/
+ #- ./workflows:/app/workflows
# - ./workflows/ocrd_workflows:/app/workflows/ocrd_workflows
# - TODO/custom/data
\ No newline at end of file
diff --git a/prepare.sh b/prepare.sh
new file mode 100644
index 0000000000000000000000000000000000000000..cd60dcfa37f82000f312481a917e4ff01a0e6467
--- /dev/null
+++ b/prepare.sh
@@ -0,0 +1,41 @@
+#!/bin/bash
+
+mkdir gt
+
+while IFS= read -r URL; do
+ OWNER=$(echo "$URL" | cut -d'/' -f4)
+ REPO=$(echo "$URL" | cut -d'/' -f5)
+ if [[ ! -f gt/"$REPO".zip ]]; then
+ echo "Downloading $REPO …"
+ RESULT=$(curl -L \
+ -H "Accept: application/vnd.github+json" \
+ -H "X-GitHub-Api-Version: 2022-11-28" \
+ https://api.github.com/repos/"$OWNER"/"$REPO"/releases/latest)
+ ZIP_URL=$(echo "$RESULT" | jq -r '.assets | .[].browser_download_url')
+ curl -L -o gt/"$REPO".zip "$ZIP_URL"
+ fi
+done < default_data_sources.txt
+
+cd gt || exit
+# the default data is structured like this:
+# repository_name.zip
+# |___ subordinate_work_1.zip
+# |___ subordinate_work_2.zip
+# |___ ...
+# $ZIP refers to the release itself which is on level "repository_name.zip"
+# the subordinate works are also OCR-D BagIts / zips. these are referred to by $INNER_ZIP.
+for ZIP in *.zip; do
+ NAME=$(echo "$ZIP" | cut -d"." -f1)
+ echo "Processing $NAME"
+ unzip -qq -d "$NAME" "$ZIP"
+ mv "$NAME"/ocrdzip_out/* "$NAME" && rm -r "$NAME"/ocrdzip_out
+ for INNER_ZIP in "$NAME"/*.zip; do
+ echo "Dealing with inner zip files …"
+ INNER_ZIP_NAME=$(basename "$INNER_ZIP" .ocrd.zip)
+ unzip -qq -d "$NAME"/"$INNER_ZIP_NAME" "$INNER_ZIP" && rm "$INNER_ZIP"
+
+ echo "Done."
+ done
+done
+
+echo " … and ready to go!"
diff --git a/src/benchmark_extraction.py b/src/benchmark_extraction.py
index 728b1b965b50af7b2e1b31a5fc911ff14046839e..2dfe6c2c4d29671f20ef6de3471455adf7d9cd98 100644
--- a/src/benchmark_extraction.py
+++ b/src/benchmark_extraction.py
@@ -8,7 +8,6 @@ from os import listdir, scandir
from statistics import stdev, median
from typing import Any, Dict, List, Union
-import yaml
from .constants import METS, RESULTS, QUIVER_MAIN, OCRD
@@ -113,7 +112,7 @@ def get_gt_workspace(workspace_path: str) -> Dict[str, str]:
if split_workspace_name[1] == 'ant':
font = 'Antiqua'
elif split_workspace_name[1] == 'frak':
- font = 'Black letter'
+ font = 'Fraktur'
else:
font = 'Font Mix'
url = 'https://github.com/OCR-D/quiver-data/blob/main/' + workspace_name_wo_workflow + '.ocrd.zip'
@@ -134,15 +133,15 @@ def get_document_metadata(workspace_path: str) -> Dict[str, Dict[str, str]]:
'layout': ''
}
}
- with open(workspace_path + '/METADATA.yml', 'r', encoding='utf-8') as file:
- metadata = yaml.safe_load(file)
+ with open(workspace_path + '/metadata.json', 'r', encoding='utf-8') as file:
+ metadata = json.load(file)
scripts = metadata['script']
fonts = []
for script in scripts:
if script == 'Latn':
fonts.append('Antiqua')
if script == 'Goth':
- fonts.append('Black Letter')
+ fonts.append('Fraktur')
if script == 'Hebr':
fonts.append('Hebrew')
if script == 'Grek':
diff --git a/workflows/execute_workflows.sh b/workflows/execute_workflows.sh
index 0429460a9093a61e89ebf397133692ac7cb709da..5f3f59f70675edac6b98004ad885d0beb49b2bdf 100755
--- a/workflows/execute_workflows.sh
+++ b/workflows/execute_workflows.sh
@@ -5,19 +5,129 @@ WORKFLOW_DIR="$ROOT"/workflows
OCRD_WORKFLOW_DIR="$WORKFLOW_DIR"/ocrd_workflows
WORKSPACE_DIR="$WORKFLOW_DIR"/workspaces
-if [[ -d workflows/workspaces ]]; then
+set -euo pipefail
+
+clean_up_dirs() {
+ if [[ -d workflows/workspaces ]]; then
rm -rf workflows/workspaces
-fi
+ fi
-if [[ -d workflows/nf-results ]]; then
- rm -rf workflows/nf-results
-fi
+ if [[ -d workflows/nf-results ]]; then
+ rm -rf workflows/nf-results
+ fi
-if [[ -d workflows/results ]]; then
- rm -rf workflows/results
-fi
+ if [[ -d workflows/results ]]; then
+ rm -rf workflows/results
+ fi
+}
-set -euo pipefail
+convert_ocrd_wfs_to_NextFlow() {
+ cd "$OCRD_WORKFLOW_DIR" || exit
+
+ echo "Convert OCR-D workflows to NextFlow …"
+
+ mkdir -p "$WORKFLOW_DIR/nf-results"
+
+ for FILE in *.txt
+ do
+ oton convert -I "$FILE" -O "$FILE".nf
+ # the venv part is not needed since we execute this in an image derived from ocrd/all:maximum
+ sed -i 's/source "${params.venv_path}"//g' "$FILE".nf
+ sed -i 's/deactivate//g' "$FILE".nf
+ done
+}
+
+download_models() {
+ echo "Download the necessary models if not available"
+ if [[ ! -f /usr/local/share/ocrd-resources/ocrd-tesserocr-recognize/ ]]
+ then
+ mkdir -p /usr/local/share/ocrd-resources/
+ ocrd resmgr download ocrd-tesserocr-recognize '*'
+ fi
+ if [[ ! -d /usr/local/share/ocrd-resources/ocrd-calamari-recognize/qurator-gt4histocr-1.0 ]]
+ then
+ mkdir -p /usr/local/share/ocrd-resources/
+ ocrd resmgr download ocrd-calamari-recognize qurator-gt4histocr-1.0
+ fi
+}
+
+create_wf_specific_workspaces() {
+ # execute this workflow on the existing data (incl. evaluation)
+ mkdir -p "$WORKSPACE_DIR"/tmp
+ cd "$WORKSPACE_DIR" || exit
+
+ # create workspace for all OCR workflows.
+ # each workflow has a separate workspace to work with.
+ echo "Create workflow specific workspaces for each dir in ./gt …"
+ for DIR in "$ROOT"/gt/*/; do
+ DIR_NAME=$(basename "$DIR")
+ if grep -q "multivolume work" <<< "$(cat $DIR/mets.xml)"; then
+ echo "$DIR_NAME is a multivolume work"
+
+ for WORKFLOW in "$OCRD_WORKFLOW_DIR"/*ocr.txt.nf
+ do
+ WF_NAME=$(basename -s .txt.nf "$WORKFLOW")
+ for SUB_WORK in $DIR/*/; do
+ SUB_WORK_DIR_NAME=$(basename "$SUB_WORK")
+ TARGET="$WORKSPACE_DIR"/tmp/"$DIR_NAME"_"$SUB_WORK_DIR_NAME"_"$WF_NAME"
+ cp -r "$ROOT"/gt/"$DIR_NAME"/"$SUB_WORK_DIR_NAME" "$TARGET"
+ if [[ -f "$ROOT"/gt/"$DIR_NAME"/metadata.json ]]; then
+ cp -r "$ROOT"/gt/"$DIR_NAME"/metadata.json "$TARGET"/metadata.json
+ fi
+ cp "$WORKFLOW" "$TARGET"/data/
+ done
+
+ done
+ else
+ for WORKFLOW in "$OCRD_WORKFLOW_DIR"/*ocr.txt.nf
+ do
+ WF_NAME=$(basename -s .txt.nf "$WORKFLOW")
+ cp -r "$ROOT"/gt/"$DIR_NAME" "$WORKSPACE_DIR"/tmp/"$DIR_NAME"_"$WF_NAME"
+ cp "$WORKFLOW" "$WORKSPACE_DIR"/tmp/"$DIR_NAME"_"$WF_NAME"/
+ done
+ fi
+ done
+}
+
+clean_up_tmp_dirs() {
+ echo "Clean up intermediate dirs …"
+ for DIR in "$WORKSPACE_DIR"/tmp/*
+ do
+ DIR_NAME=$(basename "$DIR")
+ mv "$DIR" "$WORKSPACE_DIR"/"$DIR_NAME"
+ cp "$OCRD_WORKFLOW_DIR"/*eval.txt.nf "$WORKSPACE_DIR"/"$DIR_NAME"
+ ls "$WORKSPACE_DIR"/"$DIR_NAME"
+ cp -r "$WORKSPACE_DIR"/"$DIR_NAME"/data/* "$WORKSPACE_DIR"/"$DIR_NAME"/
+ done
+
+ rm -rf "$WORKSPACE_DIR"/tmp
+ rm -rf "$WORKSPACE_DIR"/log.log
+}
+
+execute_wfs_and_extract_benchmarks() {
+ mkdir -p "$ROOT"/workflows/results
+ # for all data sets…
+ for WS_DIR in "$WORKSPACE_DIR"/*
+ do
+ if [ -d "$WS_DIR" ]; then
+ echo "Switching to $WS_DIR."
+
+ DIR_NAME=$(basename $WS_DIR)
+
+ run "$WS_DIR"/*ocr.txt.nf "$DIR_NAME" "$WS_DIR"
+ run "$WS_DIR"/*eval.txt.nf "$DIR_NAME" "$WS_DIR"
+
+ # create a result JSON according to the specs
+ echo "Get Benchmark JSON …"
+ quiver benchmarks-extraction "$WS_DIR" "$WORKFLOW"
+ echo "Done."
+
+ # move data to results dir
+ mv "$WS_DIR"/*.json "$WORKFLOW_DIR"/results
+ fi
+ done
+ cd "$ROOT" || exit
+}
adjust_workflow_settings() {
# $1: $WORKFLOW
@@ -68,103 +178,29 @@ save_workspaces() {
mv "$WORKSPACE_DIR"/"$2".zip "$WORKFLOW_DIR"/results/"$2"_"$WORKFLOW_NAME".zip
}
-cd "$OCRD_WORKFLOW_DIR" || exit
-
-echo "Convert OCR-D workflows to NextFlow …"
-
-mkdir -p "$WORKFLOW_DIR/nf-results"
-
-for FILE in *.txt
-do
- oton convert -I "$FILE" -O "$FILE".nf
- # the venv part is not needed since we execute this in an image derived from ocrd/all:maximum
- sed -i 's/source "${params.venv_path}"//g' "$FILE".nf
- sed -i 's/deactivate//g' "$FILE".nf
-done
-
-# download the necessary models if not available
-echo "Download the necessary models if not available"
-if [[ ! -f /usr/local/share/ocrd-resources/ocrd-tesserocr-recognize/ ]]
-then
- mkdir -p /usr/local/share/ocrd-resources/
- ocrd resmgr download ocrd-tesserocr-recognize '*'
-fi
-if [[ ! -d /usr/local/share/ocrd-resources/ocrd-calamari-recognize/qurator-gt4histocr-1.0 ]]
-then
- mkdir -p /usr/local/share/ocrd-resources/
- ocrd resmgr download ocrd-calamari-recognize qurator-gt4histocr-1.0
-fi
-
-# execute this workflow on the existing data (incl. evaluation)
-mkdir -p "$WORKSPACE_DIR"/tmp
-cd "$WORKSPACE_DIR" || exit
-
-# create workspace for all OCR workflows.
-# each workflow has a separate workspace to work with.
-
-echo "Restore OCR-D workspaces from BagIts and create workflow specific workspaces …"
-for BAGIT in "$ROOT"/submodules/quiver-data/*.zip
-do
- BAGIT_NAME=$(basename -s .ocrd.zip "$BAGIT")
- ocrd zip spill "$BAGIT" -d "$WORKSPACE_DIR"/tmp > "$WORKSPACE_DIR"/log.log
- unzip "$BAGIT" METADATA.yml -d "$WORKSPACE_DIR"/tmp/"$BAGIT_NAME"
-
- for WORKFLOW in "$OCRD_WORKFLOW_DIR"/*ocr.txt.nf
- do
- WF_NAME=$(basename -s .txt.nf "$WORKFLOW")
- cp -r "$WORKSPACE_DIR"/tmp/"$BAGIT_NAME" "$WORKSPACE_DIR"/tmp/"$BAGIT_NAME"_"$WF_NAME"
- cp "$WORKFLOW" "$WORKSPACE_DIR"/tmp/"$BAGIT_NAME"_"$WF_NAME"/
- done
- rm -r "$WORKSPACE_DIR"/tmp/"$BAGIT_NAME"
-done
-
-echo "Clean up intermediate dirs …"
-for DIR in "$WORKSPACE_DIR"/tmp/*
-do
- DIR_NAME=$(basename "$DIR")
- mv "$DIR" "$WORKSPACE_DIR"/"$DIR_NAME"
- cp "$OCRD_WORKFLOW_DIR"/*eval.txt.nf "$WORKSPACE_DIR"/"$DIR_NAME"
-done
-
-rm -rf "$WORKSPACE_DIR"/tmp
-rm -rf "$WORKSPACE_DIR"/log.log
-
-# start webserver for evaluation
-uvicorn api:app --app-dir "$ROOT"/src &
-
-mkdir -p "$ROOT"/workflows/results
-
-# for all data sets…
-for WS_DIR in "$WORKSPACE_DIR"/*
-do
- if [ -d "$WS_DIR" ]; then
- echo "Switching to $WS_DIR."
-
- DIR_NAME=$(basename $WS_DIR)
-
- run "$WS_DIR"/*ocr.txt.nf "$DIR_NAME" "$WS_DIR"
- run "$WS_DIR"/*eval.txt.nf "$DIR_NAME" "$WS_DIR"
-
- # create a result JSON according to the specs
- echo "Get Benchmark JSON …"
- quiver benchmarks-extraction "$WS_DIR" "$WORKFLOW"
- echo "Done."
-
- # move data to results dir
- mv "$WS_DIR"/*.json "$WORKFLOW_DIR"/results
- fi
-done
+summarize_to_data_json() {
+ # summarize JSONs
+ echo "Summarize JSONs to one file …"
+ quiver summarize-benchmarks
+ echo "Done."
+}
-cd "$ROOT" || exit
-# summarize JSONs
-echo "Summarize JSONs to one file …"
-quiver summarize-benchmarks
-echo "Done."
+final_clean_up() {
+ echo "Cleaning up …"
+ rm -rf "$WORKSPACE_DIR"
+ rm -rf "$ROOT"/work
+ rm -rf "$WORKFLOW_DIR"/nf-results
+ rm -rf "$WORKFLOW_DIR"/results
+ rm "$WORKFLOW_DIR"/ocrd_workflows/*.nf
+}
-echo "Cleaning up …"
-rm -rf "$WORKSPACE_DIR"
-rm -rf "$ROOT"/work
-rm -rf "$WORKFLOW_DIR"/nf-results
-rm -rf "$WORKFLOW_DIR"/results
-rm "$WORKFLOW_DIR"/ocrd_workflows/*.nf
+clean_up_dirs
+convert_ocrd_wfs_to_NextFlow
+download_models
+create_wf_specific_workspaces
+clean_up_tmp_dirs
+uvicorn api:app --app-dir "$ROOT"/src & # start webserver for evaluation
+execute_wfs_and_extract_benchmarks
+summarize_to_data_json
+final_clean_up