From 3a2c7272762d481a7a6ec22265b5d6e7e96bdb8b Mon Sep 17 00:00:00 2001
From: mweidling <13831557+mweidling@users.noreply.github.com>
Date: Wed, 19 Apr 2023 12:06:55 +0200
Subject: [PATCH] Add reichsanzeiger gt (#5)
* feat: add Reichsanzeiger-GT to prepare.sh
* add Reichsanzeiger subsets
* add download for Reichsanzeiger subsets
* add first draft for reichsanzeiger gt
* extract sample bundles
* update .gitignore
* clean up
* tidy up project root
* tidy up volumes
* update README
* update data
* build: roll back to version where cis runs still runs
* build: remove default model mounting
---
.gitignore | 3 +-
Dockerfile | 7 +-
README.md | 2 +-
data/workflows.json | 8857 +++--------------
.../default_data_sources.txt | 0
data_srcs/reichsanzeiger_full.txt | 1 +
data_srcs/reichsanzeiger_many_ads.list | 5 +
data_srcs/reichsanzeiger_random.list | 6 +
data_srcs/reichsanzeiger_tables.list | 5 +
data_srcs/reichsanzeiger_title_pages.list | 5 +
docker-compose.yml | 5 +-
prepare.sh | 41 -
scripts/convert-yml-to-json.py | 22 +
scripts/prepare.sh | 98 +
scripts/prepare_reichsanzeiger_sets.sh | 68 +
src/benchmark_extraction.py | 18 +-
workflows/execute_workflows.sh | 54 +-
17 files changed, 1542 insertions(+), 7655 deletions(-)
rename default_data_sources.txt => data_srcs/default_data_sources.txt (100%)
create mode 100644 data_srcs/reichsanzeiger_full.txt
create mode 100644 data_srcs/reichsanzeiger_many_ads.list
create mode 100644 data_srcs/reichsanzeiger_random.list
create mode 100644 data_srcs/reichsanzeiger_tables.list
create mode 100644 data_srcs/reichsanzeiger_title_pages.list
delete mode 100644 prepare.sh
create mode 100644 scripts/convert-yml-to-json.py
create mode 100644 scripts/prepare.sh
create mode 100644 scripts/prepare_reichsanzeiger_sets.sh
diff --git a/.gitignore b/.gitignore
index 77ec93a..5b5e2ef 100644
--- a/.gitignore
+++ b/.gitignore
@@ -6,7 +6,8 @@ workflows/workspaces
work/
workflows/nf-results/*
workflows/results
-workflows/ocrd-workflows/*.nf
+workflows/ocrd_workflows/*.nf
models
.idea
gt/*
+build/
diff --git a/Dockerfile b/Dockerfile
index ae906df..fe94011 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -1,4 +1,4 @@
-FROM ocrd/all:maximum
+FROM ocrd/all:2023-02-07
WORKDIR /app
@@ -12,6 +12,8 @@ RUN apt-get install -y --fix-missing openjdk-11-jre
COPY src src
COPY setup.py setup.py
COPY README.md README.md
+COPY scripts scripts
+COPY data_srcs data_srcs
RUN git init
RUN git submodule add https://github.com/MehmedGIT/OtoN_Converter submodules/oton
@@ -22,9 +24,6 @@ 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
diff --git a/README.md b/README.md
index 98e95f0..d065e1f 100644
--- a/README.md
+++ b/README.md
@@ -26,7 +26,7 @@ Otherwise, the tool will download all `ocrd-tesserocr-recognize` models as well
- (optional) [customize](#custom-workflows-and-data) QuiVer Benchmarks according to your needs
- 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 scripts/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
diff --git a/data/workflows.json b/data/workflows.json
index b377f0b..bbc48ae 100644
--- a/data/workflows.json
+++ b/data/workflows.json
@@ -1,7 +1,7 @@
[
{
- "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",
+ "eval_workflow_id": "wf-datareichsanzeiger_tables-eval",
+ "label": "Workflow on data reichsanzeiger_tables",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -12,16 +12,16 @@
"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/reichsanzeiger_tables.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_tables"
},
"ocr_workspace": {
- "@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"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_tables_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_tables"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_complex_alberti_pictura_1540_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_ant_complex_alberti_pictura_1540_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_tables_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_tables"
},
"workflow_steps": [
{
@@ -56,5863 +56,62 @@
"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": 7.386463,
- "cpu_time": 9.83532,
- "cer_mean": 0.10240852523716282,
- "cer_median": 0.10536980749746708,
- "cer_range": [
- 0.07124352331606218,
- 0.1306122448979592
- ],
- "cer_standard_deviation": 0.02979493530847308,
- "wer": 0.23466068901129858,
- "pages_per_minute": 24.368902951250146
- },
- "by_page": [
- {
- "page_id": "phys_0007",
- "cer": 0.07124352331606218,
- "wer": 0.2231404958677686
- },
- {
- "page_id": "phys_0008",
- "cer": 0.10536980749746708,
- "wer": 0.2484472049689441
- },
- {
- "page_id": "phys_0009",
- "cer": 0.1306122448979592,
- "wer": 0.2323943661971831
- }
- ]
- }
- },
- {
- "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",
- "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_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_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/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": [
- {
- "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": 1,
- "layout": "simple"
- }
- }
- },
- "evaluation_results": {
- "document_wide": {
- "wall_time": 8.055356999999999,
- "cpu_time": 12.105611,
- "cer_mean": 0.0199501246882793,
- "cer_median": 0.0199501246882793,
- "cer_range": [
- 0.0199501246882793,
- 0.0199501246882793
- ],
- "cer_standard_deviation": null,
- "wer": 0.09836065573770492,
- "pages_per_minute": 7.44845945375233
- },
- "by_page": [
- {
- "page_id": "phys_0001",
- "cer": 0.0199501246882793,
- "wer": 0.09836065573770492
- }
- ]
- }
- },
- {
- "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/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_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/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-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": 76.547632,
- "cpu_time": 113.81064699999999,
- "cer_mean": 0.05959514724403742,
- "cer_median": 0.0324919239726542,
- "cer_range": [
- 0.008839779005524863,
- 0.16455696202531644
- ],
- "cer_standard_deviation": 0.07087591559854228,
- "wer": 0.13238323035014285,
- "pages_per_minute": 3.1353027354262246
- },
- "by_page": [
- {
- "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-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",
- "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_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/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": [
- {
- "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": 154.933488,
- "cpu_time": 215.310644,
- "cer_mean": 0.12432674382712249,
- "cer_median": 0.12040572654031183,
- "cer_range": [
- 0.0718294051627385,
- 0.18466611706512778
- ],
- "cer_standard_deviation": 0.04665572742682036,
- "wer": 0.18974770097390478,
- "pages_per_minute": 1.5490518098966441
- },
- "by_page": [
- {
- "page_id": "phys_0001",
- "cer": 0.12757201646090535,
- "wer": 0.14035087719298245
- },
- {
- "page_id": "phys_0002",
- "cer": 0.1132394366197183,
- "wer": 0.1950354609929078
- },
- {
- "page_id": "phys_0003",
- "cer": 0.0718294051627385,
- "wer": 0.15087719298245614
- },
- {
- "page_id": "phys_0004",
- "cer": 0.18466611706512778,
- "wer": 0.2727272727272727
- }
- ]
- }
- },
- {
- "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",
- "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"
- },
- "ocr_workspace": {
- "@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/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": [
- {
- "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": 67.309474,
- "cpu_time": 101.85330800000001,
- "cer_mean": 0.16726583056278752,
- "cer_median": 0.09637318392327315,
- "cer_range": [
- 0.03187250996015936,
- 0.4444444444444444
- ],
- "cer_standard_deviation": 0.18889822286887584,
- "wer": 0.28912998545359864,
- "pages_per_minute": 3.565619900699269
- },
- "by_page": [
- {
- "page_id": "phys_0001",
- "cer": 0.4444444444444444,
- "wer": 0.6204379562043796
- },
- {
- "page_id": "phys_0002",
- "cer": 0.03187250996015936,
- "wer": 0.10919540229885058
- },
- {
- "page_id": "phys_0003",
- "cer": 0.1266778523489933,
- "wer": 0.2681564245810056
- },
- {
- "page_id": "phys_0004",
- "cer": 0.06606851549755302,
- "wer": 0.15873015873015872
- }
- ]
- }
- },
- {
- "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/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_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_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-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": 122.82167999999999,
- "cpu_time": 173.67148600000002,
- "cer_mean": 0.05991629368071161,
- "cer_median": 0.0496849827491332,
- "cer_range": [
- 0.0019973368841544607,
- 0.13829787234042554
- ],
- "cer_standard_deviation": 0.06406814571829976,
- "wer": 0.09724037283059882,
- "pages_per_minute": 1.954052411593784
- },
- "by_page": [
- {
- "page_id": "phys_0001",
- "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_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",
- "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_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_complex_luz_blitz_1784_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_frak_complex_luz_blitz_1784_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": 145.735792,
- "cpu_time": 208.46528999999998,
- "cer_mean": 0.05256131079939043,
- "cer_median": 0.029898228835687546,
- "cer_range": [
- 0.009184845005740528,
- 0.1412639405204461
- ],
- "cer_standard_deviation": 0.06044359233653714,
- "wer": 0.15936235092429332,
- "pages_per_minute": 1.6468157664384877
- },
- "by_page": [
- {
- "page_id": "phys_0001",
- "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-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",
- "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_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/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": [
- {
- "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": 36.087631,
- "cpu_time": 47.946267999999996,
- "cer_mean": 0.5161379724470825,
- "cer_median": 0.5161379724470825,
- "cer_range": [
- 0.0754414125200642,
- 0.9568345323741008
- ],
- "cer_standard_deviation": 0.6232390519399567,
- "wer": 0.6192780337941628,
- "pages_per_minute": 3.3252390548994475
- },
- "by_page": [
- {
- "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_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/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_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_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": [
- {
- "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": "complex"
- }
- }
- },
- "evaluation_results": {
- "document_wide": {
- "wall_time": 34.033219,
- "cpu_time": 53.635996999999996,
- "cer_mean": 0.06688670097710152,
- "cer_median": 0.03746177370030581,
- "cer_range": [
- 0.0290519877675841,
- 0.13414634146341464
- ],
- "cer_standard_deviation": 0.058400133164399716,
- "wer": 0.12383493220296966,
- "pages_per_minute": 5.2889501871686
- },
- "by_page": [
- {
- "page_id": "phys_0001",
- "cer": 0.13414634146341464,
- "wer": 0.22058823529411764
- },
- {
- "page_id": "phys_0002",
- "cer": 0.0290519877675841,
- "wer": 0.05357142857142857
- },
- {
- "page_id": "phys_0003",
- "cer": 0.03746177370030581,
- "wer": 0.09734513274336283
- }
- ]
- }
- },
- {
- "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/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_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/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-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": 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"
+ "Antiqua"
],
- "publication_century": "1500-1600",
+ "publication_century": "1820-1939",
"publication_decade": "",
- "publication_year": "16th century",
- "number_of_pages": 3,
- "layout": "complex"
+ "publication_year": "19th century",
+ "number_of_pages": 5,
+ "layout": "reichsanzeiger-gt"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 69.338131,
- "cpu_time": 122.13893500000002,
- "cer_mean": 0.11095656062741122,
- "cer_median": 0.12244897959183673,
+ "wall_time": 530.564159,
+ "cpu_time": 724.869668,
+ "cer_mean": 1.479972651687655,
+ "cer_median": 1.331401349741961,
"cer_range": [
- 0.07700101317122594,
- 0.133419689119171
+ 1.090097148229395,
+ 1.9778823058446757
],
- "cer_standard_deviation": 0.02991360090611332,
- "wer": 0.26341315349276156,
- "pages_per_minute": 2.595974212226747
+ "cer_standard_deviation": 0.3854879135899152,
+ "wer": 1.7910299442503619,
+ "pages_per_minute": 0.5654358571175178
},
"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
- }
+ "page_id": "P_1877_7_0059",
+ "cer": 1.090097148229395,
+ "wer": 1.4218399401645474
},
{
- "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
- }
+ "page_id": "P_1883_55_0044",
+ "cer": 1.2073319135990264,
+ "wer": 1.4705882352941178
},
{
- "id": "ocrd-cis-ocropy-dewarp",
- "params": {
- "dpi": 0,
- "range": 4.0,
- "smoothness": 1.0,
- "max_neighbour": 0.05
- }
+ "page_id": "P_1929_250_0019",
+ "cer": 1.331401349741961,
+ "wer": 1.7407221664994985
},
{
- "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": "P_1932_300_0488",
+ "cer": 1.9778823058446757,
+ "wer": 2.349493487698987
},
{
- "page_id": "phys_0002",
- "cer": 0.04932735426008968,
- "wer": 0.21621621621621623
+ "page_id": "P_1936_123_0292",
+ "cer": 1.7931505410232167,
+ "wer": 1.9725058915946583
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-datasilesius_seelenlust01_1657-eval",
+ "label": "Workflow on data silesius_seelenlust01_1657",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -5923,16 +122,16 @@
"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/silesius_seelenlust01_1657.ocrd.zip",
+ "label": "GT workspace silesius_seelenlust01_1657"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_trota_mordtbrenner_1540_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 16_frak_simple_trota_mordtbrenner_1540_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/silesius_seelenlust01_1657_ocr.zip",
+ "label": "OCR workspace for silesius_seelenlust01_1657"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_frak_simple_trota_mordtbrenner_1540_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_frak_simple_trota_mordtbrenner_1540_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/silesius_seelenlust01_1657_evaluation.zip",
+ "label": "Evaluation workspace for silesius_seelenlust01_1657"
},
"workflow_steps": [
{
@@ -5968,385 +167,279 @@
"data_properties": {
"fonts": [
"Antiqua",
- "Black Letter"
+ "Fraktur"
],
- "publication_century": "1500-1600",
+ "publication_century": "1600-1700",
"publication_decade": "",
- "publication_year": "16th century",
- "number_of_pages": 2,
- "layout": "simple"
+ "publication_year": "17th century",
+ "number_of_pages": 5,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 7.257076,
- "cpu_time": 8.706393,
- "cer_mean": 0.043071000855431994,
- "cer_median": 0.043071000855431994,
+ "wall_time": 13.583684,
+ "cpu_time": 17.039027,
+ "cer_mean": 0.3007055286105995,
+ "cer_median": 0.2951219512195122,
"cer_range": [
- 0.014285714285714285,
- 0.0718562874251497
+ 0.19271623672230653,
+ 0.44970414201183434
],
- "cer_standard_deviation": 0.04070854266369089,
- "wer": 0.10714905431199029,
- "pages_per_minute": 16.53558540657422
+ "cer_standard_deviation": 0.10657123719012947,
+ "wer": 0.5174305966287508,
+ "pages_per_minute": 22.08531941703002
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.0718562874251497,
- "wer": 0.19469026548672566
+ "cer": 0.44970414201183434,
+ "wer": 0.7101449275362319
},
{
"page_id": "phys_0002",
- "cer": 0.014285714285714285,
- "wer": 0.0196078431372549
+ "cer": 0.20913884007029876,
+ "wer": 0.5104166666666666
+ },
+ {
+ "page_id": "phys_0003",
+ "cer": 0.2951219512195122,
+ "wer": 0.4647887323943662
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.19271623672230653,
+ "wer": 0.3870967741935484
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 0.35684647302904565,
+ "wer": 0.5147058823529411
}
]
}
},
{
- "eval_workflow_id": "wf-data18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr-eval",
- "label": "Workflow on data 18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr",
+ "eval_workflow_id": "wf-databenner_herrnhuterey04_1748-eval",
+ "label": "Workflow on data benner_herrnhuterey04_1748",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
- "label": "OCR Workflow selected_pages_ocr"
+ "@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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/benner_herrnhuterey04_1748.ocrd.zip",
+ "label": "GT workspace benner_herrnhuterey04_1748"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/benner_herrnhuterey04_1748_ocr.zip",
+ "label": "OCR workspace for benner_herrnhuterey04_1748"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_ant_simple_ballenstedt_delatio_1777_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/benner_herrnhuterey04_1748_evaluation.zip",
+ "label": "Evaluation workspace for benner_herrnhuterey04_1748"
},
"workflow_steps": [
{
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
- "params": {
- "dpi": 0,
- "rulerRatioMax": 50.0,
- "rulerRatioMin": 3.0,
- "rulerAreaMax": 0.3,
- "rulerAreaMin": 0.01,
- "rulerWidthMax": 0.95,
- "columnAreaMin": 0.05,
- "columnSepWidthMax": 0.04,
- "marginTop": 0.25,
- "marginBottom": 0.75,
- "marginLeft": 0.3,
- "marginRight": 0.7,
- "padding": 10
- }
- },
- {
- "id": "ocrd-skimage-binarize",
- "params": {
- "method": "li",
- "level-of-operation": "page",
- "dpi": 0,
- "window_size": 301,
- "k": 0.34
- }
- },
- {
- "id": "ocrd-skimage-denoise",
- "params": {
- "level-of-operation": "page",
- "dpi": 0,
- "protect": 0.0,
- "maxsize": 1.0
- }
- },
- {
- "id": "ocrd-tesserocr-deskew",
- "params": {
- "operation_level": "page",
- "dpi": 0,
- "min_orientation_confidence": 1.5
- }
- },
- {
- "id": "ocrd-cis-ocropy-segment",
- "params": {
- "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"
+ "Fraktur"
],
"publication_century": "1700-1800",
"publication_decade": "",
"publication_year": "18th century",
- "number_of_pages": 3,
- "layout": "simple"
+ "number_of_pages": 4,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 189.99469000000002,
- "cpu_time": 255.53890399999997,
- "cer_mean": 0.03947936370768366,
- "cer_median": 0.0376249265138154,
+ "wall_time": 19.971749,
+ "cpu_time": 23.245253,
+ "cer_mean": 0.20433470587543373,
+ "cer_median": 0.1700173533179325,
"cer_range": [
- 0.022336769759450172,
- 0.05847639484978541
+ 0.04063745019920319,
+ 0.43666666666666665
],
- "cer_standard_deviation": 0.01814103989293333,
- "wer": 0.15140780173936938,
- "pages_per_minute": 0.947394898246893
+ "cer_standard_deviation": 0.19097881579976753,
+ "wer": 0.31449770246107817,
+ "pages_per_minute": 12.016974577439363
},
"by_page": [
{
- "page_id": "phys_00003",
- "cer": 0.022336769759450172,
- "wer": 0.10970464135021098
+ "page_id": "phys_0001",
+ "cer": 0.43666666666666665,
+ "wer": 0.583941605839416
},
{
- "page_id": "phys_00005",
- "cer": 0.0376249265138154,
- "wer": 0.16942148760330578
+ "page_id": "phys_0002",
+ "cer": 0.04063745019920319,
+ "wer": 0.10344827586206896
},
{
- "page_id": "phys_00010",
- "cer": 0.05847639484978541,
- "wer": 0.17509727626459143
+ "page_id": "phys_0003",
+ "cer": 0.05536912751677853,
+ "wer": 0.18435754189944134
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.28466557911908646,
+ "wer": 0.3862433862433862
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-databohse_helicon_1696-eval",
+ "label": "Workflow on data bohse_helicon_1696",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
- "label": "OCR Workflow selected_pages_ocr"
- },
- "eval_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
- "label": "Evaluation Workflow dinglehopper_eval"
- },
- "gt_workspace": {
- "@id": "https://github.com/OCR-D/quiver-data/blob/main/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": "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/bohse_helicon_1696.ocrd.zip",
+ "label": "GT workspace bohse_helicon_1696"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bohse_helicon_1696_ocr.zip",
+ "label": "OCR workspace for bohse_helicon_1696"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bohse_helicon_1696_evaluation.zip",
+ "label": "Evaluation workspace for bohse_helicon_1696"
+ },
+ "workflow_steps": [
{
- "id": "ocrd-cis-ocropy-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"
+ "Fraktur"
],
"publication_century": "1600-1700",
"publication_decade": "",
"publication_year": "17th century",
- "number_of_pages": 3,
+ "number_of_pages": 5,
"layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 88.38021099999999,
- "cpu_time": 127.13157100000001,
- "cer_mean": 0.047703401203510305,
- "cer_median": 0.04723502304147465,
+ "wall_time": 16.040643,
+ "cpu_time": 19.265802,
+ "cer_mean": 0.40403175304113403,
+ "cer_median": 0.45517241379310347,
"cer_range": [
- 0.04592720970537262,
- 0.04994797086368366
+ 0.1955040871934605,
+ 0.5079365079365079
],
- "cer_standard_deviation": 0.002050893378518156,
- "wer": 0.17277993391305632,
- "pages_per_minute": 2.0366550154536296
+ "cer_standard_deviation": 0.12207840266111031,
+ "wer": 0.5356334009868995,
+ "pages_per_minute": 18.70249216318822
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.04994797086368366,
- "wer": 0.21019108280254778
+ "cer": 0.5079365079365079,
+ "wer": 0.6382978723404256
},
{
"page_id": "phys_0002",
- "cer": 0.04723502304147465,
- "wer": 0.15172413793103448
+ "cer": 0.40540540540540543,
+ "wer": 0.5740740740740741
},
{
"page_id": "phys_0003",
- "cer": 0.04592720970537262,
- "wer": 0.1564245810055866
+ "cer": 0.45517241379310347,
+ "wer": 0.5467625899280576
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.45614035087719296,
+ "wer": 0.6018518518518519
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 0.1955040871934605,
+ "wer": 0.31718061674008813
}
]
}
},
{
- "eval_workflow_id": "wf-data18_frak_complex_buerger_gedichte_1778_minimal_ocr-eval",
- "label": "Workflow on data 18_frak_complex_buerger_gedichte_1778_minimal_ocr",
+ "eval_workflow_id": "wf-databuerger_gedichte_1778-eval",
+ "label": "Workflow on data buerger_gedichte_1778",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -6357,16 +450,16 @@
"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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/buerger_gedichte_1778.ocrd.zip",
+ "label": "GT workspace buerger_gedichte_1778"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_buerger_gedichte_1778_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 18_frak_complex_buerger_gedichte_1778_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/buerger_gedichte_1778_ocr.zip",
+ "label": "OCR workspace for buerger_gedichte_1778"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_buerger_gedichte_1778_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_frak_complex_buerger_gedichte_1778_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/buerger_gedichte_1778_evaluation.zip",
+ "label": "Evaluation workspace for buerger_gedichte_1778"
},
"workflow_steps": [
{
@@ -6402,7 +495,7 @@
"data_properties": {
"fonts": [
"Antiqua",
- "Black Letter"
+ "Fraktur"
],
"publication_century": "1700-1800",
"publication_decade": "",
@@ -6414,8 +507,8 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 6.016163,
- "cpu_time": 7.291227,
+ "wall_time": 5.921714,
+ "cpu_time": 7.016172,
"cer_mean": 0.10672693293515115,
"cer_median": 0.10672693293515115,
"cer_range": [
@@ -6424,7 +517,7 @@
],
"cer_standard_deviation": 0.08751702749046443,
"wer": 0.23135901859306116,
- "pages_per_minute": 19.94626807817541
+ "pages_per_minute": 20.26440317786371
},
"by_page": [
{
@@ -6441,8 +534,8 @@
}
},
{
- "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",
+ "eval_workflow_id": "wf-datarollenhagen_reysen_1603-eval",
+ "label": "Workflow on data rollenhagen_reysen_1603",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -6453,16 +546,16 @@
"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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/rollenhagen_reysen_1603.ocrd.zip",
+ "label": "GT workspace rollenhagen_reysen_1603"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_ant_simple_ballenstedt_delatio_1777_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 18_ant_simple_ballenstedt_delatio_1777_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/rollenhagen_reysen_1603_ocr.zip",
+ "label": "OCR workspace for rollenhagen_reysen_1603"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_ant_simple_ballenstedt_delatio_1777_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_ant_simple_ballenstedt_delatio_1777_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/rollenhagen_reysen_1603_evaluation.zip",
+ "label": "Evaluation workspace for rollenhagen_reysen_1603"
},
"workflow_steps": [
{
@@ -6498,11 +591,11 @@
"data_properties": {
"fonts": [
"Antiqua",
- "Black Letter"
+ "Fraktur"
],
- "publication_century": "1700-1800",
+ "publication_century": "1600-1700",
"publication_decade": "",
- "publication_year": "18th century",
+ "publication_year": "17th century",
"number_of_pages": 3,
"layout": "simple"
}
@@ -6510,40 +603,40 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 24.13474,
- "cpu_time": 27.011507,
- "cer_mean": 0.06438361522903834,
- "cer_median": 0.03969957081545064,
+ "wall_time": 16.777821,
+ "cpu_time": 19.16547,
+ "cer_mean": 0.21046220070684576,
+ "cer_median": 0.16929133858267717,
"cer_range": [
- 0.021764032073310423,
- 0.13168724279835392
+ 0.14512471655328799,
+ 0.3169705469845722
],
- "cer_standard_deviation": 0.058972490200809365,
- "wer": 0.16906902212925057,
- "pages_per_minute": 7.458128821773095
+ "cer_standard_deviation": 0.093027024434784,
+ "wer": 0.3190752126565147,
+ "pages_per_minute": 10.728449183001775
},
"by_page": [
{
- "page_id": "phys_00003",
- "cer": 0.021764032073310423,
- "wer": 0.12236286919831224
+ "page_id": "phys_0001",
+ "cer": 0.3169705469845722,
+ "wer": 0.4649122807017544
},
{
- "page_id": "phys_00005",
- "cer": 0.13168724279835392,
- "wer": 0.2603305785123967
+ "page_id": "phys_0002",
+ "cer": 0.16929133858267717,
+ "wer": 0.28104575163398693
},
{
- "page_id": "phys_00010",
- "cer": 0.03969957081545064,
- "wer": 0.1245136186770428
+ "page_id": "phys_0003",
+ "cer": 0.14512471655328799,
+ "wer": 0.2112676056338028
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-dataeuler_rechenkunst01_1738-eval",
+ "label": "Workflow on data euler_rechenkunst01_1738",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -6554,16 +647,16 @@
"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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/euler_rechenkunst01_1738.ocrd.zip",
+ "label": "GT workspace euler_rechenkunst01_1738"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 17_frak_complex_huebner_handbuch_1696_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/euler_rechenkunst01_1738_ocr.zip",
+ "label": "OCR workspace for euler_rechenkunst01_1738"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 17_frak_complex_huebner_handbuch_1696_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/euler_rechenkunst01_1738_evaluation.zip",
+ "label": "Evaluation workspace for euler_rechenkunst01_1738"
},
"workflow_steps": [
{
@@ -6599,134 +692,200 @@
"data_properties": {
"fonts": [
"Antiqua",
- "Black Letter"
+ "Fraktur"
],
- "publication_century": "1600-1700",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "17th century",
- "number_of_pages": 3,
+ "publication_year": "18th century",
+ "number_of_pages": 6,
"layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 16.954402,
- "cpu_time": 19.205537,
- "cer_mean": 0.20627969569747484,
- "cer_median": 0.08702290076335878,
+ "wall_time": 18.514995,
+ "cpu_time": 23.577633,
+ "cer_mean": 0.25835849983393794,
+ "cer_median": 0.230420483908856,
"cer_range": [
- 0.08487084870848709,
- 0.44694533762057875
+ 0.08586296617519515,
+ 0.5747368421052632
],
- "cer_standard_deviation": 0.20842533731220508,
- "wer": 0.3271033769383102,
- "pages_per_minute": 10.616711813250621
+ "cer_standard_deviation": 0.16915619139008603,
+ "wer": 0.37827288927088376,
+ "pages_per_minute": 19.443699552713895
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.44694533762057875,
- "wer": 0.6341463414634146
+ "cer": 0.08586296617519515,
+ "wer": 0.10465116279069768
},
{
"page_id": "phys_0002",
- "cer": 0.08702290076335878,
- "wer": 0.1504424778761062
+ "cer": 0.5747368421052632,
+ "wer": 0.7987804878048781
},
{
"page_id": "phys_0003",
- "cer": 0.08487084870848709,
- "wer": 0.19672131147540983
+ "cer": 0.2767102229054573,
+ "wer": 0.3786407766990291
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.22828282828282828,
+ "wer": 0.27741935483870966
+ },
+ {
+ "page_id": "phys_0005",
+ "cer": 0.152,
+ "wer": 0.27218934911242604
+ },
+ {
+ "page_id": "phys_0006",
+ "cer": 0.23255813953488372,
+ "wer": 0.43795620437956206
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-datakistler_kraeuter_1500-eval",
+ "label": "Workflow on data kistler_kraeuter_1500",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
- "label": "OCR Workflow slower_processors_ocr"
+ "@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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/kistler_kraeuter_1500.ocrd.zip",
+ "label": "GT workspace kistler_kraeuter_1500"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_arnimb_goethe03_1835_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 19_frak_simple_arnimb_goethe03_1835_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/kistler_kraeuter_1500_ocr.zip",
+ "label": "OCR workspace for kistler_kraeuter_1500"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_frak_simple_arnimb_goethe03_1835_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 19_frak_simple_arnimb_goethe03_1835_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/kistler_kraeuter_1500_evaluation.zip",
+ "label": "Evaluation workspace for kistler_kraeuter_1500"
},
"workflow_steps": [
{
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
+ "id": "ocrd-tesserocr-recognize",
"params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
"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
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
}
- },
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 7.833456,
+ "cpu_time": 9.538595,
+ "cer_mean": 0.1809783660358373,
+ "cer_median": 0.1809783660358373,
+ "cer_range": [
+ 0.13479623824451412,
+ 0.2271604938271605
+ ],
+ "cer_standard_deviation": 0.06531139146173669,
+ "wer": 0.408373786407767,
+ "pages_per_minute": 15.318909048573197
+ },
+ "by_page": [
{
- "id": "ocrd-cis-ocropy-denoise",
- "params": {
- "level-of-operation": "page",
- "noise_maxsize": 3.0,
- "dpi": 0
- }
+ "page_id": "phys_0007",
+ "cer": 0.2271604938271605,
+ "wer": 0.441747572815534
},
{
- "id": "ocrd-tesserocr-deskew",
- "params": {
- "operation_level": "page",
- "dpi": 0,
- "min_orientation_confidence": 1.5
- }
- },
+ "page_id": "phys_0021",
+ "cer": 0.13479623824451412,
+ "wer": 0.375
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datacalvi_beutelschneider01_1627-eval",
+ "label": "Workflow on data calvi_beutelschneider01_1627",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/calvi_beutelschneider01_1627.ocrd.zip",
+ "label": "GT workspace calvi_beutelschneider01_1627"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/calvi_beutelschneider01_1627_ocr.zip",
+ "label": "OCR workspace for calvi_beutelschneider01_1627"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/calvi_beutelschneider01_1627_evaluation.zip",
+ "label": "Evaluation workspace for calvi_beutelschneider01_1627"
+ },
+ "workflow_steps": [
{
- "id": "ocrd-tesserocr-segment",
+ "id": "ocrd-tesserocr-recognize",
"params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
"dpi": 0,
- "padding": 4,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
"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": "",
@@ -6737,29 +896,95 @@
"auto_model": false,
"oem": "DEFAULT"
}
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 7.37275,
+ "cpu_time": 9.77562,
+ "cer_mean": 0.08254540344028655,
+ "cer_median": 0.09331797235023041,
+ "cer_range": [
+ 0.05025996533795494,
+ 0.1040582726326743
+ ],
+ "cer_standard_deviation": 0.02847104928930229,
+ "wer": 0.1821206964037463,
+ "pages_per_minute": 24.41422806958055
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.1040582726326743,
+ "wer": 0.21656050955414013
},
{
- "id": "ocrd-cis-ocropy-dewarp",
- "params": {
- "dpi": 0,
- "range": 4.0,
- "smoothness": 1.0,
- "max_neighbour": 0.05
- }
+ "page_id": "phys_0002",
+ "cer": 0.09331797235023041,
+ "wer": 0.20689655172413793
},
+ {
+ "page_id": "phys_0003",
+ "cer": 0.05025996533795494,
+ "wer": 0.12290502793296089
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-dataarnimb_goethe03_1835-eval",
+ "label": "Workflow on data arnimb_goethe03_1835",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/arnimb_goethe03_1835.ocrd.zip",
+ "label": "GT workspace arnimb_goethe03_1835"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/arnimb_goethe03_1835_ocr.zip",
+ "label": "OCR workspace for arnimb_goethe03_1835"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/arnimb_goethe03_1835_evaluation.zip",
+ "label": "Evaluation workspace for arnimb_goethe03_1835"
+ },
+ "workflow_steps": [
{
"id": "ocrd-tesserocr-recognize",
"params": {
- "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,
@@ -6779,7 +1004,7 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Black Letter"
+ "Fraktur"
],
"publication_century": "1800-1900",
"publication_decade": "",
@@ -6791,112 +1016,67 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 12.164379999999998,
- "cpu_time": 19.670232,
- "cer_mean": 0.0056657223796034,
- "cer_median": 0.0056657223796034,
+ "wall_time": 4.707975,
+ "cpu_time": 5.426059,
+ "cer_mean": 0.004721435316336166,
+ "cer_median": 0.004721435316336166,
"cer_range": [
- 0.0056657223796034,
- 0.0056657223796034
+ 0.004721435316336166,
+ 0.004721435316336166
],
"cer_standard_deviation": null,
- "wer": 0.031746031746031744,
- "pages_per_minute": 4.932433876613524
+ "wer": 0.015873015873015872,
+ "pages_per_minute": 12.744332754528221
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.0056657223796034,
- "wer": 0.031746031746031744
+ "cer": 0.004721435316336166,
+ "wer": 0.015873015873015872
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-datablumenbach_anatomie_1805-eval",
+ "label": "Workflow on data blumenbach_anatomie_1805",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
- "label": "OCR Workflow slower_processors_ocr"
+ "@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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/blumenbach_anatomie_1805.ocrd.zip",
+ "label": "GT workspace blumenbach_anatomie_1805"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 17_frak_complex_silesius_seelenlust01_1657_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/blumenbach_anatomie_1805_ocr.zip",
+ "label": "OCR workspace for blumenbach_anatomie_1805"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 17_frak_complex_silesius_seelenlust01_1657_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/blumenbach_anatomie_1805_evaluation.zip",
+ "label": "Evaluation workspace for blumenbach_anatomie_1805"
},
"workflow_steps": [
{
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
- "params": {
- "dpi": 0,
- "rulerRatioMax": 50.0,
- "rulerRatioMin": 3.0,
- "rulerAreaMax": 0.3,
- "rulerAreaMin": 0.01,
- "rulerWidthMax": 0.95,
- "columnAreaMin": 0.05,
- "columnSepWidthMax": 0.04,
- "marginTop": 0.25,
- "marginBottom": 0.75,
- "marginLeft": 0.3,
- "marginRight": 0.7,
- "padding": 10
- }
- },
- {
- "id": "ocrd-cis-ocropy-denoise",
- "params": {
- "level-of-operation": "page",
- "noise_maxsize": 3.0,
- "dpi": 0
- }
- },
- {
- "id": "ocrd-tesserocr-deskew",
- "params": {
- "operation_level": "page",
- "dpi": 0,
- "min_orientation_confidence": 1.5
- }
- },
- {
- "id": "ocrd-tesserocr-segment",
+ "id": "ocrd-tesserocr-recognize",
"params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
"dpi": 0,
- "padding": 4,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
"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": "",
@@ -6907,29 +1087,95 @@
"auto_model": false,
"oem": "DEFAULT"
}
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1800-1900",
+ "publication_decade": "",
+ "publication_year": "19th century",
+ "number_of_pages": 3,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 13.404813,
+ "cpu_time": 15.520204,
+ "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": 13.428012759297722
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.08736842105263158,
+ "wer": 0.22666666666666666
},
{
- "id": "ocrd-cis-ocropy-dewarp",
- "params": {
- "dpi": 0,
- "range": 4.0,
- "smoothness": 1.0,
- "max_neighbour": 0.05
- }
+ "page_id": "phys_0002",
+ "cer": 0.04055496264674493,
+ "wer": 0.14814814814814814
},
+ {
+ "page_id": "phys_0003",
+ "cer": 0.12192262602579132,
+ "wer": 0.33076923076923076
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datareichsanzeiger_title_pages-eval",
+ "label": "Workflow on data reichsanzeiger_title_pages",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_title_pages.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_title_pages"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_title_pages_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_title_pages"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_title_pages_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_title_pages"
+ },
+ "workflow_steps": [
{
"id": "ocrd-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,
@@ -6949,63 +1195,62 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua",
- "Black Letter"
+ "Antiqua"
],
- "publication_century": "1600-1700",
+ "publication_century": "1820-1939",
"publication_decade": "",
- "publication_year": "17th century",
+ "publication_year": "19th century",
"number_of_pages": 5,
- "layout": "complex"
+ "layout": "reichsanzeiger-gt"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 51.051955,
- "cpu_time": 79.096845,
- "cer_mean": 0.16456973001701142,
- "cer_median": 0.11229135053110774,
+ "wall_time": 470.537927,
+ "cpu_time": 603.540066,
+ "cer_mean": 0.5074594248191785,
+ "cer_median": 0.3389393598117977,
"cer_range": [
- 0.05917159763313609,
- 0.34146341463414637
+ 0.09154875911296263,
+ 1.0
],
- "cer_standard_deviation": 0.11554307430913685,
- "wer": 0.34312458192490203,
- "pages_per_minute": 5.876366536795701
+ "cer_standard_deviation": 0.36783069273412317,
+ "wer": 0.630009265947987,
+ "pages_per_minute": 0.6375681593037663
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.05917159763313609,
- "wer": 0.2463768115942029
+ "page_id": "P_1881_115_0163",
+ "cer": 0.3377672361293631,
+ "wer": 0.4791666666666667
},
{
- "page_id": "phys_0002",
- "cer": 0.0913884007029877,
- "wer": 0.28125
+ "page_id": "P_1885_5_0054",
+ "cer": 0.09154875911296263,
+ "wer": 0.20966135458167331
},
{
- "page_id": "phys_0003",
- "cer": 0.34146341463414637,
- "wer": 0.5774647887323944
+ "page_id": "P_1887_134_0444",
+ "cer": 1.0,
+ "wer": 1.0
},
{
- "page_id": "phys_0004",
- "cer": 0.11229135053110774,
- "wer": 0.1693548387096774
+ "page_id": "P_1916_169_0087",
+ "cer": 0.3389393598117977,
+ "wer": 0.5072414380644062
},
{
- "page_id": "phys_0005",
- "cer": 0.21853388658367912,
- "wer": 0.4411764705882353
+ "page_id": "P_1918_267_0129",
+ "cer": 0.769041769041769,
+ "wer": 0.953976870427189
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-dataluther_auszlegunge_1520-eval",
+ "label": "Workflow on data luther_auszlegunge_1520",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -7016,16 +1261,16 @@
"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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/luther_auszlegunge_1520.ocrd.zip",
+ "label": "GT workspace luther_auszlegunge_1520"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 17_frak_complex_silesius_seelenlust01_1657_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luther_auszlegunge_1520_ocr.zip",
+ "label": "OCR workspace for luther_auszlegunge_1520"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 17_frak_complex_silesius_seelenlust01_1657_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luther_auszlegunge_1520_evaluation.zip",
+ "label": "Evaluation workspace for luther_auszlegunge_1520"
},
"workflow_steps": [
{
@@ -7061,144 +1306,84 @@
"data_properties": {
"fonts": [
"Antiqua",
- "Black Letter"
+ "Fraktur"
],
- "publication_century": "1600-1700",
+ "publication_century": "1500-1600",
"publication_decade": "",
- "publication_year": "17th century",
- "number_of_pages": 5,
- "layout": "complex"
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 13.146098,
- "cpu_time": 16.982357,
- "cer_mean": 0.3024629978548877,
- "cer_median": 0.2951219512195122,
+ "wall_time": 10.160861,
+ "cpu_time": 11.973182,
+ "cer_mean": 0.24855845660550213,
+ "cer_median": 0.24855845660550213,
"cer_range": [
- 0.19271623672230653,
- 0.44970414201183434
+ 0.07865168539325842,
+ 0.4184652278177458
],
- "cer_standard_deviation": 0.10474043149462715,
- "wer": 0.5195139299620842,
- "pages_per_minute": 22.820459728810782
+ "cer_standard_deviation": 0.2402844601873776,
+ "wer": 0.37300307219662054,
+ "pages_per_minute": 11.81002279235982
},
"by_page": [
- {
- "page_id": "phys_0001",
- "cer": 0.44970414201183434,
- "wer": 0.7101449275362319
- },
- {
- "page_id": "phys_0002",
- "cer": 0.2179261862917399,
- "wer": 0.5208333333333334
- },
{
"page_id": "phys_0003",
- "cer": 0.2951219512195122,
- "wer": 0.4647887323943662
- },
- {
- "page_id": "phys_0004",
- "cer": 0.19271623672230653,
- "wer": 0.3870967741935484
+ "cer": 0.4184652278177458,
+ "wer": 0.48333333333333334
},
{
- "page_id": "phys_0005",
- "cer": 0.35684647302904565,
- "wer": 0.5147058823529411
+ "page_id": "phys_0029",
+ "cer": 0.07865168539325842,
+ "wer": 0.2626728110599078
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-datahuebner_handbuch_1696-eval",
+ "label": "Workflow on data huebner_handbuch_1696",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
- "label": "OCR Workflow slower_processors_ocr"
+ "@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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/huebner_handbuch_1696.ocrd.zip",
+ "label": "GT workspace huebner_handbuch_1696"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_luz_blitz_1784_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 18_frak_complex_luz_blitz_1784_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/huebner_handbuch_1696_ocr.zip",
+ "label": "OCR workspace for huebner_handbuch_1696"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_luz_blitz_1784_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_frak_complex_luz_blitz_1784_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/huebner_handbuch_1696_evaluation.zip",
+ "label": "Evaluation workspace for huebner_handbuch_1696"
},
"workflow_steps": [
{
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
- "params": {
- "dpi": 0,
- "rulerRatioMax": 50.0,
- "rulerRatioMin": 3.0,
- "rulerAreaMax": 0.3,
- "rulerAreaMin": 0.01,
- "rulerWidthMax": 0.95,
- "columnAreaMin": 0.05,
- "columnSepWidthMax": 0.04,
- "marginTop": 0.25,
- "marginBottom": 0.75,
- "marginLeft": 0.3,
- "marginRight": 0.7,
- "padding": 10
- }
- },
- {
- "id": "ocrd-cis-ocropy-denoise",
- "params": {
- "level-of-operation": "page",
- "noise_maxsize": 3.0,
- "dpi": 0
- }
- },
- {
- "id": "ocrd-tesserocr-deskew",
- "params": {
- "operation_level": "page",
- "dpi": 0,
- "min_orientation_confidence": 1.5
- }
- },
- {
- "id": "ocrd-tesserocr-segment",
+ "id": "ocrd-tesserocr-recognize",
"params": {
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
"dpi": 0,
- "padding": 4,
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
"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": "",
@@ -7209,29 +1394,95 @@
"auto_model": false,
"oem": "DEFAULT"
}
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1600-1700",
+ "publication_decade": "",
+ "publication_year": "17th century",
+ "number_of_pages": 3,
+ "layout": "complex"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 15.981865,
+ "cpu_time": 18.106327,
+ "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": 11.262765640931143
+ },
+ "by_page": [
+ {
+ "page_id": "phys_0001",
+ "cer": 0.44694533762057875,
+ "wer": 0.6341463414634146
},
{
- "id": "ocrd-cis-ocropy-dewarp",
- "params": {
- "dpi": 0,
- "range": 4.0,
- "smoothness": 1.0,
- "max_neighbour": 0.05
- }
+ "page_id": "phys_0002",
+ "cer": 0.08702290076335878,
+ "wer": 0.1504424778761062
},
+ {
+ "page_id": "phys_0003",
+ "cer": 0.08487084870848709,
+ "wer": 0.19672131147540983
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datann_besuch_1780-eval",
+ "label": "Workflow on data nn_besuch_1780",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/nn_besuch_1780.ocrd.zip",
+ "label": "GT workspace nn_besuch_1780"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/nn_besuch_1780_ocr.zip",
+ "label": "OCR workspace for nn_besuch_1780"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/nn_besuch_1780_evaluation.zip",
+ "label": "Evaluation workspace for nn_besuch_1780"
+ },
+ "workflow_steps": [
{
"id": "ocrd-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,
@@ -7252,7 +1503,7 @@
"data_properties": {
"fonts": [
"Antiqua",
- "Black Letter"
+ "Fraktur"
],
"publication_century": "1700-1800",
"publication_decade": "",
@@ -7264,173 +1515,105 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 64.58104600000001,
- "cpu_time": 99.479389,
- "cer_mean": 0.02547899799369996,
- "cer_median": 0.027204160076875337,
+ "wall_time": 14.316583,
+ "cpu_time": 16.701302,
+ "cer_mean": 0.06315407734401027,
+ "cer_median": 0.026726016076928857,
"cer_range": [
- 0.010332950631458095,
- 0.03717472118959108
+ 0.01878453038674033,
+ 0.18037974683544303
],
- "cer_standard_deviation": 0.011266997500136374,
- "wer": 0.07714749104131867,
- "pages_per_minute": 3.716260650222357
+ "cer_standard_deviation": 0.07825196427362012,
+ "wer": 0.11662846322503488,
+ "pages_per_minute": 16.76377666374721
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.03717472118959108,
- "wer": 0.10256410256410256
+ "page_id": "phys_00001",
+ "cer": 0.18037974683544303,
+ "wer": 0.1896551724137931
},
{
- "page_id": "phys_0002",
- "cer": 0.025119617224880382,
- "wer": 0.09565217391304348
+ "page_id": "phys_00002",
+ "cer": 0.01878453038674033,
+ "wer": 0.08670520231213873
},
{
- "page_id": "phys_0003",
- "cer": 0.029288702928870293,
- "wer": 0.07924528301886792
+ "page_id": "phys_00003",
+ "cer": 0.02505446623093682,
+ "wer": 0.08670520231213873
},
{
- "page_id": "phys_0004",
- "cer": 0.010332950631458095,
- "wer": 0.0311284046692607
+ "page_id": "phys_00004",
+ "cer": 0.028397565922920892,
+ "wer": 0.10344827586206896
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-databallenstedt_delatio_1777-eval",
+ "label": "Workflow on data ballenstedt_delatio_1777",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/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/19_ant_simple.ocrd.zip",
- "label": "GT workspace 19th century Antiqua simple layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/ballenstedt_delatio_1777.ocrd.zip",
+ "label": "GT workspace ballenstedt_delatio_1777"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_blumenbach_anatomie_1805_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 19_ant_simple_blumenbach_anatomie_1805_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/ballenstedt_delatio_1777_ocr.zip",
+ "label": "OCR workspace for ballenstedt_delatio_1777"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/19_ant_simple_blumenbach_anatomie_1805_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 19_ant_simple_blumenbach_anatomie_1805_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/ballenstedt_delatio_1777_evaluation.zip",
+ "label": "Evaluation workspace for ballenstedt_delatio_1777"
},
"workflow_steps": [
{
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
- "params": {
- "dpi": 0,
- "rulerRatioMax": 50.0,
- "rulerRatioMin": 3.0,
- "rulerAreaMax": 0.3,
- "rulerAreaMin": 0.01,
- "rulerWidthMax": 0.95,
- "columnAreaMin": 0.05,
- "columnSepWidthMax": 0.04,
- "marginTop": 0.25,
- "marginBottom": 0.75,
- "marginLeft": 0.3,
- "marginRight": 0.7,
- "padding": 10
- }
- },
- {
- "id": "ocrd-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"
+ "Fraktur"
],
- "publication_century": "1800-1900",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "19th century",
+ "publication_year": "18th century",
"number_of_pages": 3,
"layout": "simple"
}
@@ -7438,155 +1621,75 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 97.867795,
- "cpu_time": 142.045967,
- "cer_mean": 0.08457767160660273,
- "cer_median": 0.07737397420867527,
+ "wall_time": 21.444519,
+ "cpu_time": 24.028484,
+ "cer_mean": 0.06438361522903834,
+ "cer_median": 0.03969957081545064,
"cer_range": [
- 0.0416221985058698,
- 0.13473684210526315
+ 0.021764032073310423,
+ 0.13168724279835392
],
- "cer_standard_deviation": 0.046973440956685215,
- "wer": 0.3081861348528015,
- "pages_per_minute": 1.8392158523649174
+ "cer_standard_deviation": 0.058972490200809365,
+ "wer": 0.16906902212925057,
+ "pages_per_minute": 8.393753200992757
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.13473684210526315,
- "wer": 0.38666666666666666
+ "page_id": "phys_00003",
+ "cer": 0.021764032073310423,
+ "wer": 0.12236286919831224
},
{
- "page_id": "phys_0002",
- "cer": 0.0416221985058698,
- "wer": 0.21481481481481482
+ "page_id": "phys_00005",
+ "cer": 0.13168724279835392,
+ "wer": 0.2603305785123967
},
{
- "page_id": "phys_0003",
- "cer": 0.07737397420867527,
- "wer": 0.3230769230769231
+ "page_id": "phys_00010",
+ "cer": 0.03969957081545064,
+ "wer": 0.1245136186770428
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-datareichsanzeiger_many_ads-eval",
+ "label": "Workflow on data reichsanzeiger_many_ads",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/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_frak_complex.ocrd.zip",
- "label": "GT workspace 18th century Black letter complex layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_many_ads.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_many_ads"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_nn_besuch_1780_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 18_frak_complex_nn_besuch_1780_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_many_ads_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_many_ads"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_nn_besuch_1780_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_frak_complex_nn_besuch_1780_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_many_ads_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_many_ads"
},
"workflow_steps": [
- {
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
- "params": {
- "dpi": 0,
- "rulerRatioMax": 50.0,
- "rulerRatioMin": 3.0,
- "rulerAreaMax": 0.3,
- "rulerAreaMin": 0.01,
- "rulerWidthMax": 0.95,
- "columnAreaMin": 0.05,
- "columnSepWidthMax": 0.04,
- "marginTop": 0.25,
- "marginBottom": 0.75,
- "marginLeft": 0.3,
- "marginRight": 0.7,
- "padding": 10
- }
- },
- {
- "id": "ocrd-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,
@@ -7606,173 +1709,97 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua",
- "Black Letter"
+ "Antiqua"
],
- "publication_century": "1700-1800",
+ "publication_century": "1820-1939",
"publication_decade": "",
- "publication_year": "18th century",
- "number_of_pages": 4,
- "layout": "complex"
+ "publication_year": "19th century",
+ "number_of_pages": 5,
+ "layout": "reichsanzeiger-gt"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 49.367179,
- "cpu_time": 71.436347,
- "cer_mean": 0.07900170040044713,
- "cer_median": 0.018462174141687326,
+ "wall_time": 344.194331,
+ "cpu_time": 401.989378,
+ "cer_mean": 1.3340240550306781,
+ "cer_median": 0.7412443874278384,
"cer_range": [
- 0.013259668508287293,
- 0.26582278481012656
+ 0.3960932753867003,
+ 3.7402255639097746
],
- "cer_standard_deviation": 0.1245848990441196,
- "wer": 0.13667696498571524,
- "pages_per_minute": 4.8615295599531825
+ "cer_standard_deviation": 1.378192503303974,
+ "wer": 1.8653811990174032,
+ "pages_per_minute": 0.8716006423708356
},
"by_page": [
{
- "page_id": "phys_00001",
- "cer": 0.26582278481012656,
- "wer": 0.3620689655172414
+ "page_id": "P_1871_155_0279",
+ "cer": 0.3960932753867003,
+ "wer": 0.49564980967917344
},
{
- "page_id": "phys_00002",
- "cer": 0.013259668508287293,
- "wer": 0.057803468208092484
+ "page_id": "P_1871_65_0045",
+ "cer": 0.7412443874278384,
+ "wer": 0.947814451382694
},
{
- "page_id": "phys_00003",
- "cer": 0.020697167755991286,
- "wer": 0.06936416184971098
+ "page_id": "P_1873_1_0017",
+ "cer": 0.5852251348300515,
+ "wer": 0.6788418708240535
},
{
- "page_id": "phys_00004",
- "cer": 0.016227180527383367,
- "wer": 0.05747126436781609
+ "page_id": "P_1881_1_0662",
+ "cer": 3.7402255639097746,
+ "wer": 5.734011627906977
+ },
+ {
+ "page_id": "P_1883_55_0044",
+ "cer": 1.2073319135990264,
+ "wer": 1.4705882352941178
}
]
}
},
{
- "eval_workflow_id": "wf-data17_frak_complex_huebner_handbuch_1696_slower_processors_ocr-eval",
- "label": "Workflow on data 17_frak_complex_huebner_handbuch_1696_slower_processors_ocr",
+ "eval_workflow_id": "wf-databernd_lebensbeschreibung_1738-eval",
+ "label": "Workflow on data bernd_lebensbeschreibung_1738",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/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/17_frak_complex.ocrd.zip",
- "label": "GT workspace 17th century Black letter complex layout"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/bernd_lebensbeschreibung_1738.ocrd.zip",
+ "label": "GT workspace bernd_lebensbeschreibung_1738"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 17_frak_complex_huebner_handbuch_1696_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bernd_lebensbeschreibung_1738_ocr.zip",
+ "label": "OCR workspace for bernd_lebensbeschreibung_1738"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 17_frak_complex_huebner_handbuch_1696_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/bernd_lebensbeschreibung_1738_evaluation.zip",
+ "label": "Evaluation workspace for bernd_lebensbeschreibung_1738"
},
"workflow_steps": [
- {
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
- "params": {
- "dpi": 0,
- "rulerRatioMax": 50.0,
- "rulerRatioMin": 3.0,
- "rulerAreaMax": 0.3,
- "rulerAreaMin": 0.01,
- "rulerWidthMax": 0.95,
- "columnAreaMin": 0.05,
- "columnSepWidthMax": 0.04,
- "marginTop": 0.25,
- "marginBottom": 0.75,
- "marginLeft": 0.3,
- "marginRight": 0.7,
- "padding": 10
- }
- },
- {
- "id": "ocrd-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,
@@ -7793,11 +1820,11 @@
"data_properties": {
"fonts": [
"Antiqua",
- "Black Letter"
+ "Fraktur"
],
- "publication_century": "1600-1700",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "17th century",
+ "publication_year": "18th century",
"number_of_pages": 3,
"layout": "complex"
}
@@ -7805,40 +1832,40 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 61.992684,
- "cpu_time": 83.585045,
- "cer_mean": 0.20641950275814583,
- "cer_median": 0.1099236641221374,
+ "wall_time": 8.577653,
+ "cpu_time": 10.790615,
+ "cer_mean": 0.1790109644215708,
+ "cer_median": 0.05504587155963303,
"cer_range": [
- 0.09132841328413284,
- 0.4180064308681672
+ 0.008409785932721712,
+ 0.4735772357723577
],
- "cer_standard_deviation": 0.18347538513029096,
- "wer": 0.3802990111723255,
- "pages_per_minute": 2.903568427526061
+ "cer_standard_deviation": 0.2561653709691831,
+ "wer": 0.2391410103864555,
+ "pages_per_minute": 20.984761216150854
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.4180064308681672,
- "wer": 0.7073170731707317
+ "cer": 0.4735772357723577,
+ "wer": 0.5882352941176471
},
{
"page_id": "phys_0002",
- "cer": 0.1099236641221374,
- "wer": 0.24778761061946902
+ "cer": 0.05504587155963303,
+ "wer": 0.09821428571428571
},
{
"page_id": "phys_0003",
- "cer": 0.09132841328413284,
- "wer": 0.18579234972677597
+ "cer": 0.008409785932721712,
+ "wer": 0.030973451327433628
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-dataweigel_gnothi02_1618-eval",
+ "label": "Workflow on data weigel_gnothi02_1618",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -7849,16 +1876,16 @@
"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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/weigel_gnothi02_1618.ocrd.zip",
+ "label": "GT workspace weigel_gnothi02_1618"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_justi_abhandlung01_1758_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 18_frak_complex_justi_abhandlung01_1758_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/weigel_gnothi02_1618_ocr.zip",
+ "label": "OCR workspace for weigel_gnothi02_1618"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_justi_abhandlung01_1758_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_frak_complex_justi_abhandlung01_1758_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/weigel_gnothi02_1618_evaluation.zip",
+ "label": "Evaluation workspace for weigel_gnothi02_1618"
},
"workflow_steps": [
{
@@ -7894,694 +1921,385 @@
"data_properties": {
"fonts": [
"Antiqua",
- "Black Letter"
+ "Fraktur"
],
- "publication_century": "1700-1800",
+ "publication_century": "1600-1700",
"publication_decade": "",
- "publication_year": "18th century",
+ "publication_year": "17th century",
"number_of_pages": 4,
- "layout": "complex"
+ "layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 14.25996,
- "cpu_time": 17.707908,
- "cer_mean": 0.1347518651788604,
- "cer_median": 0.14344023364180397,
+ "wall_time": 27.856097,
+ "cpu_time": 31.285033,
+ "cer_mean": 0.11399763421865425,
+ "cer_median": 0.09987627660053089,
"cer_range": [
- 0.09893550407013149,
- 0.15319148936170213
+ 0.07531106745252128,
+ 0.18092691622103388
],
- "cer_standard_deviation": 0.02557054025937362,
- "wer": 0.19715777371165608,
- "pages_per_minute": 16.830341740089032
+ "cer_standard_deviation": 0.047514032403792344,
+ "wer": 0.19377061794956923,
+ "pages_per_minute": 8.615708079994121
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.13375130616509928,
- "wer": 0.2
+ "cer": 0.18092691622103388,
+ "wer": 0.3282051282051282
},
{
"page_id": "phys_0002",
- "cer": 0.15319148936170213,
- "wer": 0.22916666666666666
+ "cer": 0.07531106745252128,
+ "wer": 0.15019762845849802
},
{
"page_id": "phys_0003",
- "cer": 0.09893550407013149,
- "wer": 0.12598425196850394
+ "cer": 0.11397849462365592,
+ "wer": 0.15934065934065933
},
{
"page_id": "phys_0004",
- "cer": 0.15312916111850866,
- "wer": 0.23348017621145375
+ "cer": 0.08577405857740586,
+ "wer": 0.13733905579399142
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-datalessing_menschengeschlecht_1780-eval",
+ "label": "Workflow on data lessing_menschengeschlecht_1780",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
- "label": "OCR Workflow selected_pages_ocr"
+ "@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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/lessing_menschengeschlecht_1780.ocrd.zip",
+ "label": "GT workspace lessing_menschengeschlecht_1780"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_euler_rechenkunst01_1738_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 18_frak_complex_euler_rechenkunst01_1738_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/lessing_menschengeschlecht_1780_ocr.zip",
+ "label": "OCR workspace for lessing_menschengeschlecht_1780"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_euler_rechenkunst01_1738_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_frak_complex_euler_rechenkunst01_1738_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/lessing_menschengeschlecht_1780_evaluation.zip",
+ "label": "Evaluation workspace for lessing_menschengeschlecht_1780"
},
"workflow_steps": [
{
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
- "params": {
- "dpi": 0,
- "rulerRatioMax": 50.0,
- "rulerRatioMin": 3.0,
- "rulerAreaMax": 0.3,
- "rulerAreaMin": 0.01,
- "rulerWidthMax": 0.95,
- "columnAreaMin": 0.05,
- "columnSepWidthMax": 0.04,
- "marginTop": 0.25,
- "marginBottom": 0.75,
- "marginLeft": 0.3,
- "marginRight": 0.7,
- "padding": 10
- }
- },
- {
- "id": "ocrd-skimage-binarize",
- "params": {
- "method": "li",
- "level-of-operation": "page",
- "dpi": 0,
- "window_size": 201,
- "k": 0.34
- }
- },
- {
- "id": "ocrd-skimage-denoise",
- "params": {
- "level-of-operation": "page",
- "dpi": 0,
- "protect": 0.0,
- "maxsize": 1.0
- }
- },
- {
- "id": "ocrd-tesserocr-deskew",
- "params": {
- "operation_level": "page",
- "dpi": 0,
- "min_orientation_confidence": 1.5
- }
- },
- {
- "id": "ocrd-cis-ocropy-segment",
- "params": {
- "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"
+ "Fraktur"
],
"publication_century": "1700-1800",
"publication_decade": "",
"publication_year": "18th century",
- "number_of_pages": 6,
- "layout": "complex"
+ "number_of_pages": 1,
+ "layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 161.89715999999999,
- "cpu_time": 264.56102,
- "cer_mean": 0.1953749490770702,
- "cer_median": 0.1576184687985986,
+ "wall_time": 2.026755,
+ "cpu_time": 2.56115,
+ "cer_mean": 0.02493765586034913,
+ "cer_median": 0.02493765586034913,
"cer_range": [
- 0.09627059843885516,
- 0.3873684210526316
+ 0.02493765586034913,
+ 0.02493765586034913
],
- "cer_standard_deviation": 0.11209255429867662,
- "wer": 0.3314757763924516,
- "pages_per_minute": 2.2236338179125563
+ "cer_standard_deviation": null,
+ "wer": 0.09836065573770492,
+ "pages_per_minute": 29.60397285315689
},
"by_page": [
{
"page_id": "phys_0001",
- "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
+ "cer": 0.02493765586034913,
+ "wer": 0.09836065573770492
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-dataalberti_pictura_1540-eval",
+ "label": "Workflow on data alberti_pictura_1540",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
- "label": "OCR Workflow selected_pages_ocr"
+ "@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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/alberti_pictura_1540.ocrd.zip",
+ "label": "GT workspace alberti_pictura_1540"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_heyden_paedono_1548_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 16_ant_simple_heyden_paedono_1548_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/alberti_pictura_1540_ocr.zip",
+ "label": "OCR workspace for alberti_pictura_1540"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/16_ant_simple_heyden_paedono_1548_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 16_ant_simple_heyden_paedono_1548_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/alberti_pictura_1540_evaluation.zip",
+ "label": "Evaluation workspace for alberti_pictura_1540"
},
"workflow_steps": [
{
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
- "params": {
- "dpi": 0,
- "rulerRatioMax": 50.0,
- "rulerRatioMin": 3.0,
- "rulerAreaMax": 0.3,
- "rulerAreaMin": 0.01,
- "rulerWidthMax": 0.95,
- "columnAreaMin": 0.05,
- "columnSepWidthMax": 0.04,
- "marginTop": 0.25,
- "marginBottom": 0.75,
- "marginLeft": 0.3,
- "marginRight": 0.7,
- "padding": 10
- }
- },
- {
- "id": "ocrd-skimage-binarize",
- "params": {
- "method": "li",
- "level-of-operation": "page",
- "dpi": 0,
- "window_size": 301,
- "k": 0.34
- }
- },
- {
- "id": "ocrd-skimage-denoise",
- "params": {
- "level-of-operation": "page",
- "dpi": 0,
- "protect": 0.0,
- "maxsize": 1.0
- }
- },
- {
- "id": "ocrd-tesserocr-deskew",
- "params": {
- "operation_level": "page",
- "dpi": 0,
- "min_orientation_confidence": 1.5
- }
- },
- {
- "id": "ocrd-cis-ocropy-segment",
- "params": {
- "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"
+ "Fraktur"
],
"publication_century": "1500-1600",
"publication_decade": "",
"publication_year": "16th century",
"number_of_pages": 3,
- "layout": "simple"
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 72.48043099999998,
- "cpu_time": 107.649963,
- "cer_mean": 0.05488709037929677,
- "cer_median": 0.03488372093023256,
+ "wall_time": 7.85773,
+ "cpu_time": 9.961092,
+ "cer_mean": 0.10240852523716282,
+ "cer_median": 0.10536980749746708,
"cer_range": [
- 0.028395061728395062,
- 0.10138248847926268
+ 0.07124352331606218,
+ 0.1306122448979592
],
- "cer_standard_deviation": 0.04039668560556835,
- "wer": 0.13745817563490756,
- "pages_per_minute": 2.483428940978566
+ "cer_standard_deviation": 0.02979493530847308,
+ "wer": 0.23466068901129858,
+ "pages_per_minute": 22.907379103125205
},
"by_page": [
{
"page_id": "phys_0007",
- "cer": 0.10138248847926268,
- "wer": 0.17475728155339806
+ "cer": 0.07124352331606218,
+ "wer": 0.2231404958677686
},
{
- "page_id": "phys_0013",
- "cer": 0.03488372093023256,
- "wer": 0.13592233009708737
+ "page_id": "phys_0008",
+ "cer": 0.10536980749746708,
+ "wer": 0.2484472049689441
},
{
- "page_id": "phys_0014",
- "cer": 0.028395061728395062,
- "wer": 0.1016949152542373
+ "page_id": "phys_0009",
+ "cer": 0.1306122448979592,
+ "wer": 0.2323943661971831
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-dataheyden_paedono_1548-eval",
+ "label": "Workflow on data heyden_paedono_1548",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
- "label": "OCR Workflow selected_pages_ocr"
+ "@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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/heyden_paedono_1548.ocrd.zip",
+ "label": "GT workspace heyden_paedono_1548"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 17_frak_complex_huebner_handbuch_1696_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/heyden_paedono_1548_ocr.zip",
+ "label": "OCR workspace for heyden_paedono_1548"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_huebner_handbuch_1696_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 17_frak_complex_huebner_handbuch_1696_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/heyden_paedono_1548_evaluation.zip",
+ "label": "Evaluation workspace for heyden_paedono_1548"
},
"workflow_steps": [
{
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
- "params": {
- "dpi": 0,
- "rulerRatioMax": 50.0,
- "rulerRatioMin": 3.0,
- "rulerAreaMax": 0.3,
- "rulerAreaMin": 0.01,
- "rulerWidthMax": 0.95,
- "columnAreaMin": 0.05,
- "columnSepWidthMax": 0.04,
- "marginTop": 0.25,
- "marginBottom": 0.75,
- "marginLeft": 0.3,
- "marginRight": 0.7,
- "padding": 10
- }
- },
- {
- "id": "ocrd-skimage-binarize",
- "params": {
- "method": "li",
- "level-of-operation": "page",
- "dpi": 0,
- "window_size": 301,
- "k": 0.34
- }
- },
- {
- "id": "ocrd-skimage-denoise",
- "params": {
- "level-of-operation": "page",
- "dpi": 0,
- "protect": 0.0,
- "maxsize": 1.0
- }
- },
- {
- "id": "ocrd-tesserocr-deskew",
- "params": {
- "operation_level": "page",
- "dpi": 0,
- "min_orientation_confidence": 1.5
- }
- },
- {
- "id": "ocrd-cis-ocropy-segment",
- "params": {
- "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"
+ "Fraktur"
],
- "publication_century": "1600-1700",
+ "publication_century": "1500-1600",
"publication_decade": "",
- "publication_year": "17th century",
+ "publication_year": "16th century",
"number_of_pages": 3,
- "layout": "complex"
+ "layout": "simple"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 182.12772,
- "cpu_time": 235.173884,
- "cer_mean": 0.2294897403822962,
- "cer_median": 0.06717557251908397,
+ "wall_time": 8.303854,
+ "cpu_time": 10.292532,
+ "cer_mean": 0.07452119312897007,
+ "cer_median": 0.0629800307219662,
"cer_range": [
- 0.061808118081180814,
- 0.5594855305466238
+ 0.037037037037037035,
+ 0.12354651162790697
],
- "cer_standard_deviation": 0.2857973382248686,
- "wer": 0.36395117469236493,
- "pages_per_minute": 0.9883174291096379
+ "cer_standard_deviation": 0.044394494261965886,
+ "wer": 0.22683890077340793,
+ "pages_per_minute": 21.676681695029803
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.5594855305466238,
- "wer": 0.8292682926829268
+ "page_id": "phys_0007",
+ "cer": 0.0629800307219662,
+ "wer": 0.21359223300970873
},
{
- "page_id": "phys_0002",
- "cer": 0.06717557251908397,
- "wer": 0.11504424778761062
+ "page_id": "phys_0013",
+ "cer": 0.12354651162790697,
+ "wer": 0.33980582524271846
},
{
- "page_id": "phys_0003",
- "cer": 0.061808118081180814,
- "wer": 0.14754098360655737
+ "page_id": "phys_0014",
+ "cer": 0.037037037037037035,
+ "wer": 0.1271186440677966
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-dataluz_blitz_1784-eval",
+ "label": "Workflow on data luz_blitz_1784",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/slower_processors_ocr.txt",
- "label": "OCR Workflow slower_processors_ocr"
+ "@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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/luz_blitz_1784.ocrd.zip",
+ "label": "GT workspace luz_blitz_1784"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_calvi_beutelschneider01_1627_slower_processors_ocr_ocr.zip",
- "label": "OCR workspace for 17_frak_simple_calvi_beutelschneider01_1627_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luz_blitz_1784_ocr.zip",
+ "label": "OCR workspace for luz_blitz_1784"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_simple_calvi_beutelschneider01_1627_slower_processors_ocr_evaluation.zip",
- "label": "Evaluation workspace for 17_frak_simple_calvi_beutelschneider01_1627_slower_processors_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/luz_blitz_1784_evaluation.zip",
+ "label": "Evaluation workspace for luz_blitz_1784"
},
"workflow_steps": [
- {
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
- "params": {
- "dpi": 0,
- "rulerRatioMax": 50.0,
- "rulerRatioMin": 3.0,
- "rulerAreaMax": 0.3,
- "rulerAreaMin": 0.01,
- "rulerWidthMax": 0.95,
- "columnAreaMin": 0.05,
- "columnSepWidthMax": 0.04,
- "marginTop": 0.25,
- "marginBottom": 0.75,
- "marginLeft": 0.3,
- "marginRight": 0.7,
- "padding": 10
- }
- },
- {
- "id": "ocrd-cis-ocropy-denoise",
- "params": {
- "level-of-operation": "page",
- "noise_maxsize": 3.0,
- "dpi": 0
- }
- },
- {
- "id": "ocrd-tesserocr-deskew",
- "params": {
- "operation_level": "page",
- "dpi": 0,
- "min_orientation_confidence": 1.5
- }
- },
- {
- "id": "ocrd-tesserocr-segment",
- "params": {
- "dpi": 0,
- "padding": 4,
- "shrink_polygons": false,
- "block_polygons": false,
- "find_tables": true,
- "find_staves": false,
- "sparse_text": false,
- "overwrite_segments": true,
- "segmentation_level": "region",
- "textequiv_level": "none",
- "overwrite_text": true,
- "raw_lines": false,
- "char_whitelist": "",
- "char_blacklist": "",
- "char_unblacklist": "",
- "tesseract_parameters": {},
- "xpath_parameters": {},
- "xpath_model": {},
- "auto_model": false,
- "oem": "DEFAULT"
- }
- },
- {
- "id": "ocrd-cis-ocropy-dewarp",
- "params": {
- "dpi": 0,
- "range": 4.0,
- "smoothness": 1.0,
- "max_neighbour": 0.05
- }
- },
{
"id": "ocrd-tesserocr-recognize",
"params": {
- "textequiv_level": "glyph",
- "overwrite_segments": true,
+ "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,
@@ -8602,52 +2320,57 @@
"data_properties": {
"fonts": [
"Antiqua",
- "Black Letter"
+ "Fraktur"
],
- "publication_century": "1600-1700",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "17th century",
- "number_of_pages": 3,
- "layout": "simple"
+ "publication_year": "18th century",
+ "number_of_pages": 4,
+ "layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 57.283714,
- "cpu_time": 83.17899299999999,
- "cer_mean": 0.15501219595245352,
- "cer_median": 0.11654526534859522,
+ "wall_time": 19.577649,
+ "cpu_time": 22.415525,
+ "cer_mean": 0.030367084502750087,
+ "cer_median": 0.02381735856100435,
"cer_range": [
- 0.09618717504332755,
- 0.2523041474654378
+ 0.014354066985645933,
+ 0.05947955390334572
],
- "cer_standard_deviation": 0.08486993479509113,
- "wer": 0.360948415946103,
- "pages_per_minute": 3.142254358717034
+ "cer_standard_deviation": 0.02030808283356641,
+ "wer": 0.08722583259487592,
+ "pages_per_minute": 12.258877457655922
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.11654526534859522,
- "wer": 0.46496815286624205
+ "cer": 0.05947955390334572,
+ "wer": 0.1794871794871795
},
{
"page_id": "phys_0002",
- "cer": 0.2523041474654378,
- "wer": 0.4
+ "cer": 0.014354066985645933,
+ "wer": 0.043478260869565216
},
{
"page_id": "phys_0003",
- "cer": 0.09618717504332755,
- "wer": 0.21787709497206703
+ "cer": 0.02869097429766886,
+ "wer": 0.07924528301886792
+ },
+ {
+ "page_id": "phys_0004",
+ "cer": 0.01894374282433984,
+ "wer": 0.04669260700389105
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-dataestor_rechtsgelehrsamkeit02_1758-eval",
+ "label": "Workflow on data estor_rechtsgelehrsamkeit02_1758",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -8658,16 +2381,16 @@
"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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/estor_rechtsgelehrsamkeit02_1758.ocrd.zip",
+ "label": "GT workspace estor_rechtsgelehrsamkeit02_1758"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_estor_rechtsgelehrsamkeit02_1758_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 18_frak_complex_estor_rechtsgelehrsamkeit02_1758_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/estor_rechtsgelehrsamkeit02_1758_ocr.zip",
+ "label": "OCR workspace for estor_rechtsgelehrsamkeit02_1758"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_estor_rechtsgelehrsamkeit02_1758_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_frak_complex_estor_rechtsgelehrsamkeit02_1758_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/estor_rechtsgelehrsamkeit02_1758_evaluation.zip",
+ "label": "Evaluation workspace for estor_rechtsgelehrsamkeit02_1758"
},
"workflow_steps": [
{
@@ -8703,7 +2426,7 @@
"data_properties": {
"fonts": [
"Antiqua",
- "Black Letter"
+ "Fraktur"
],
"publication_century": "1700-1800",
"publication_decade": "",
@@ -8715,8 +2438,8 @@
},
"evaluation_results": {
"document_wide": {
- "wall_time": 18.505333,
- "cpu_time": 22.105623,
+ "wall_time": 19.327455,
+ "cpu_time": 22.561449,
"cer_mean": 0.11589408928020027,
"cer_median": 0.10084584323499293,
"cer_range": [
@@ -8725,7 +2448,7 @@
],
"cer_standard_deviation": 0.07259145757108061,
"wer": 0.20102650242627845,
- "pages_per_minute": 12.969234328287957
+ "pages_per_minute": 12.417568686617043
},
"by_page": [
{
@@ -8752,187 +2475,210 @@
}
},
{
- "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",
+ "eval_workflow_id": "wf-datatrota_mordtbrenner_1540-eval",
+ "label": "Workflow on data trota_mordtbrenner_1540",
"metadata": {
"ocr_workflow": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/selected_pages_ocr.txt",
- "label": "OCR Workflow selected_pages_ocr"
+ "@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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/trota_mordtbrenner_1540.ocrd.zip",
+ "label": "GT workspace trota_mordtbrenner_1540"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_selected_pages_ocr_ocr.zip",
- "label": "OCR workspace for 17_frak_complex_silesius_seelenlust01_1657_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/trota_mordtbrenner_1540_ocr.zip",
+ "label": "OCR workspace for trota_mordtbrenner_1540"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/17_frak_complex_silesius_seelenlust01_1657_selected_pages_ocr_evaluation.zip",
- "label": "Evaluation workspace for 17_frak_complex_silesius_seelenlust01_1657_selected_pages_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/trota_mordtbrenner_1540_evaluation.zip",
+ "label": "Evaluation workspace for trota_mordtbrenner_1540"
},
"workflow_steps": [
{
- "id": "ocrd-cis-ocropy-binarize",
- "params": {
- "method": "ocropy",
- "threshold": 0.5,
- "grayscale": false,
- "maxskew": 0.0,
- "noise_maxsize": 0,
- "dpi": 0,
- "level-of-operation": "page"
- }
- },
- {
- "id": "ocrd-anybaseocr-crop",
- "params": {
- "dpi": 0,
- "rulerRatioMax": 50.0,
- "rulerRatioMin": 3.0,
- "rulerAreaMax": 0.3,
- "rulerAreaMin": 0.01,
- "rulerWidthMax": 0.95,
- "columnAreaMin": 0.05,
- "columnSepWidthMax": 0.04,
- "marginTop": 0.25,
- "marginBottom": 0.75,
- "marginLeft": 0.3,
- "marginRight": 0.7,
- "padding": 10
- }
- },
- {
- "id": "ocrd-skimage-binarize",
- "params": {
- "method": "li",
- "level-of-operation": "page",
- "dpi": 0,
- "window_size": 301,
- "k": 0.34
- }
- },
- {
- "id": "ocrd-skimage-denoise",
+ "id": "ocrd-tesserocr-recognize",
"params": {
- "level-of-operation": "page",
+ "segmentation_level": "region",
+ "textequiv_level": "word",
+ "find_tables": true,
+ "model": "Fraktur_GT4HistOCR",
"dpi": 0,
- "protect": 0.0,
- "maxsize": 1.0
+ "padding": 0,
+ "overwrite_segments": false,
+ "overwrite_text": true,
+ "shrink_polygons": false,
+ "block_polygons": false,
+ "find_staves": false,
+ "sparse_text": false,
+ "raw_lines": false,
+ "char_whitelist": "",
+ "char_blacklist": "",
+ "char_unblacklist": "",
+ "tesseract_parameters": {},
+ "xpath_parameters": {},
+ "xpath_model": {},
+ "auto_model": false,
+ "oem": "DEFAULT"
}
- },
+ }
+ ],
+ "workflow_model": "Fraktur_GT4HistOCR",
+ "eval_tool": "ocrd-dinglehopper vNone",
+ "document_metadata": {
+ "data_properties": {
+ "fonts": [
+ "Antiqua",
+ "Fraktur"
+ ],
+ "publication_century": "1500-1600",
+ "publication_decade": "",
+ "publication_year": "16th century",
+ "number_of_pages": 2,
+ "layout": "simple"
+ }
+ }
+ },
+ "evaluation_results": {
+ "document_wide": {
+ "wall_time": 7.192959,
+ "cpu_time": 8.533417,
+ "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": 16.682981232063188
+ },
+ "by_page": [
{
- "id": "ocrd-tesserocr-deskew",
- "params": {
- "operation_level": "page",
- "dpi": 0,
- "min_orientation_confidence": 1.5
- }
+ "page_id": "phys_0001",
+ "cer": 0.0718562874251497,
+ "wer": 0.19469026548672566
},
{
- "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
- }
- },
+ "page_id": "phys_0002",
+ "cer": 0.014285714285714285,
+ "wer": 0.0196078431372549
+ }
+ ]
+ }
+ },
+ {
+ "eval_workflow_id": "wf-datajusti_abhandlung01_1758-eval",
+ "label": "Workflow on data justi_abhandlung01_1758",
+ "metadata": {
+ "ocr_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
+ "label": "OCR Workflow minimal_ocr"
+ },
+ "eval_workflow": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/dinglehopper_eval.txt",
+ "label": "Evaluation Workflow dinglehopper_eval"
+ },
+ "gt_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/justi_abhandlung01_1758.ocrd.zip",
+ "label": "GT workspace justi_abhandlung01_1758"
+ },
+ "ocr_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/justi_abhandlung01_1758_ocr.zip",
+ "label": "OCR workspace for justi_abhandlung01_1758"
+ },
+ "eval_workspace": {
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/justi_abhandlung01_1758_evaluation.zip",
+ "label": "Evaluation workspace for justi_abhandlung01_1758"
+ },
+ "workflow_steps": [
{
- "id": "ocrd-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"
+ "Fraktur"
],
- "publication_century": "1600-1700",
+ "publication_century": "1700-1800",
"publication_decade": "",
- "publication_year": "17th century",
- "number_of_pages": 5,
+ "publication_year": "18th century",
+ "number_of_pages": 4,
"layout": "complex"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 108.68203399999999,
- "cpu_time": 154.24126800000005,
- "cer_mean": 0.2924684311074824,
- "cer_median": 0.2583479789103691,
+ "wall_time": 13.437096,
+ "cpu_time": 16.64254,
+ "cer_mean": 0.12942563348778183,
+ "cer_median": 0.13278777025964683,
"cer_range": [
- 0.06676783004552352,
- 0.7536585365853659
+ 0.09893550407013149,
+ 0.15319148936170213
],
- "cer_standard_deviation": 0.2774715020464407,
- "wer": 0.4982391606465789,
- "pages_per_minute": 2.7603458360008246
+ "cer_standard_deviation": 0.022501323128153037,
+ "wer": 0.19483806092245076,
+ "pages_per_minute": 17.861002109384348
},
"by_page": [
{
"page_id": "phys_0001",
- "cer": 0.08481262327416174,
- "wer": 0.43478260869565216
+ "cer": 0.13375130616509928,
+ "wer": 0.2
},
{
"page_id": "phys_0002",
- "cer": 0.2583479789103691,
- "wer": 0.5625
+ "cer": 0.15319148936170213,
+ "wer": 0.22916666666666666
},
{
"page_id": "phys_0003",
- "cer": 0.7536585365853659,
- "wer": 0.9014084507042254
+ "cer": 0.09893550407013149,
+ "wer": 0.12992125984251968
},
{
"page_id": "phys_0004",
- "cer": 0.06676783004552352,
- "wer": 0.04838709677419355
- },
- {
- "page_id": "phys_0005",
- "cer": 0.2987551867219917,
- "wer": 0.5441176470588235
+ "cer": 0.1318242343541944,
+ "wer": 0.22026431718061673
}
]
}
},
{
- "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",
+ "eval_workflow_id": "wf-datareichsanzeiger_random-eval",
+ "label": "Workflow on data reichsanzeiger_random",
"metadata": {
"ocr_workflow": {
"@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/ocrd_workflows/minimal_ocr.txt",
@@ -8943,16 +2689,16 @@
"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"
+ "@id": "https://github.com/OCR-D/quiver-data/blob/main/reichsanzeiger_random.ocrd.zip",
+ "label": "GT workspace reichsanzeiger_random"
},
"ocr_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_euler_rechenkunst01_1738_minimal_ocr_ocr.zip",
- "label": "OCR workspace for 18_frak_complex_euler_rechenkunst01_1738_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_random_ocr.zip",
+ "label": "OCR workspace for reichsanzeiger_random"
},
"eval_workspace": {
- "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/18_frak_complex_euler_rechenkunst01_1738_minimal_ocr_evaluation.zip",
- "label": "Evaluation workspace for 18_frak_complex_euler_rechenkunst01_1738_minimal_ocr"
+ "@id": "https://github.com/OCR-D/quiver-back-end/blob/main/workflows/results/reichsanzeiger_random_evaluation.zip",
+ "label": "Evaluation workspace for reichsanzeiger_random"
},
"workflow_steps": [
{
@@ -8987,61 +2733,60 @@
"document_metadata": {
"data_properties": {
"fonts": [
- "Antiqua",
- "Black Letter"
+ "Antiqua"
],
- "publication_century": "1700-1800",
+ "publication_century": "1820-1939",
"publication_decade": "",
- "publication_year": "18th century",
+ "publication_year": "19th century",
"number_of_pages": 6,
- "layout": "complex"
+ "layout": "reichsanzeiger-gt"
}
}
},
"evaluation_results": {
"document_wide": {
- "wall_time": 18.674437,
- "cpu_time": 24.291901,
- "cer_mean": 0.25835849983393794,
- "cer_median": 0.230420483908856,
+ "wall_time": 636.322401,
+ "cpu_time": 852.529652,
+ "cer_mean": 1.1094634128044334,
+ "cer_median": 0.8993673021608106,
"cer_range": [
- 0.08586296617519515,
- 0.5747368421052632
+ 0.30019453260980855,
+ 1.9778823058446757
],
- "cer_standard_deviation": 0.16915619139008603,
- "wer": 0.37827288927088376,
- "pages_per_minute": 19.277689603172508
+ "cer_standard_deviation": 0.645494993792005,
+ "wer": 1.3304151477151094,
+ "pages_per_minute": 0.565750945486516
},
"by_page": [
{
- "page_id": "phys_0001",
- "cer": 0.08586296617519515,
- "wer": 0.10465116279069768
+ "page_id": "P_1879_45_0344",
+ "cer": 0.9843462873477166,
+ "wer": 1.1759884281581485
},
{
- "page_id": "phys_0002",
- "cer": 0.5747368421052632,
- "wer": 0.7987804878048781
+ "page_id": "P_1885_5_0055",
+ "cer": 0.30019453260980855,
+ "wer": 0.40127817019845274
},
{
- "page_id": "phys_0003",
- "cer": 0.2767102229054573,
- "wer": 0.3786407766990291
+ "page_id": "P_1889_1_0018",
+ "cer": 0.7868184930272782,
+ "wer": 0.9704142011834319
},
{
- "page_id": "phys_0004",
- "cer": 0.22828282828282828,
- "wer": 0.27741935483870966
+ "page_id": "P_1891_33_0452",
+ "cer": 0.8143883169739047,
+ "wer": 1.112810707456979
},
{
- "page_id": "phys_0005",
- "cer": 0.152,
- "wer": 0.27218934911242604
+ "page_id": "P_1932_300_0488",
+ "cer": 1.9778823058446757,
+ "wer": 2.349493487698987
},
{
- "page_id": "phys_0006",
- "cer": 0.23255813953488372,
- "wer": 0.43795620437956206
+ "page_id": "P_1936_123_0292",
+ "cer": 1.7931505410232167,
+ "wer": 1.9725058915946583
}
]
}
diff --git a/default_data_sources.txt b/data_srcs/default_data_sources.txt
similarity index 100%
rename from default_data_sources.txt
rename to data_srcs/default_data_sources.txt
diff --git a/data_srcs/reichsanzeiger_full.txt b/data_srcs/reichsanzeiger_full.txt
new file mode 100644
index 0000000..87ce334
--- /dev/null
+++ b/data_srcs/reichsanzeiger_full.txt
@@ -0,0 +1 @@
+https://github.com/UB-Mannheim/reichsanzeiger-gt
\ No newline at end of file
diff --git a/data_srcs/reichsanzeiger_many_ads.list b/data_srcs/reichsanzeiger_many_ads.list
new file mode 100644
index 0000000..c5ef614
--- /dev/null
+++ b/data_srcs/reichsanzeiger_many_ads.list
@@ -0,0 +1,5 @@
+053-7972/0044.jp2&CVT=jpeg 1883_55_0044.jpg
+041-7960/0662.jp2&CVT=jpeg 1881_1_0662.jpg
+102-9983/0045.jp2&CVT=jpeg 1871_65_0045.jpg
+101-9982/0017.jp2&CVT=jpeg 1873_1_0017.jpg
+003-7922/0279.jp2&CVT=jpeg 1871_155_0279.jpg
diff --git a/data_srcs/reichsanzeiger_random.list b/data_srcs/reichsanzeiger_random.list
new file mode 100644
index 0000000..a488a19
--- /dev/null
+++ b/data_srcs/reichsanzeiger_random.list
@@ -0,0 +1,6 @@
+104-8023/0452.jp2&CVT=jpeg 1891_33_0452.jpg
+090-8009/0018.jp2&CVT=jpeg 1889_1_0018.jpg
+065-7984/0055.jp2&CVT=jpeg 1885_5_0055.jpg
+093-9053/0488.jp2&CVT=jpeg 1932_300_0488.jpg
+014-8455/0292.jp2&CVT=jpeg 1936_123_0292.jpg
+035-7954/0344.jp2&CVT=jpeg 1879_45_0344.jpg
diff --git a/data_srcs/reichsanzeiger_tables.list b/data_srcs/reichsanzeiger_tables.list
new file mode 100644
index 0000000..6518229
--- /dev/null
+++ b/data_srcs/reichsanzeiger_tables.list
@@ -0,0 +1,5 @@
+093-9053/0488.jp2&CVT=jpeg 1932_300_0488.jpg
+014-8455/0292.jp2&CVT=jpeg 1936_123_0292.jpg
+076-9036/0019.jp2&CVT=jpeg 1929_250_0019.jpg
+027-7946/0059.jp2&CVT=jpeg 1877_7_0059.jpg
+053-7972/0044.jp2&CVT=jpeg 1883_55_0044.jpg
diff --git a/data_srcs/reichsanzeiger_title_pages.list b/data_srcs/reichsanzeiger_title_pages.list
new file mode 100644
index 0000000..3fd2509
--- /dev/null
+++ b/data_srcs/reichsanzeiger_title_pages.list
@@ -0,0 +1,5 @@
+044-7963/0163.jp2&CVT=jpeg 1881_115_0163.jpg
+079-7998/0444.jp2&CVT=jpeg 1887_134_0444.jpg
+153-9561/0087.jp2&CVT=jpeg 1916_169_0087.jpg
+167-9449/0129.jp2&CVT=jpeg 1918_267_0129.jpg
+065-7984/0054.jp2&CVT=jpeg 1885_5_0054.jpg
diff --git a/docker-compose.yml b/docker-compose.yml
index c4be3ef..b7ca20e 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -10,7 +10,6 @@ services:
- ./data:/app/data # this will write the results to your host system
- ./gt:/app/gt
# mount your modules, custom workflows and data here
- - ./models:/usr/local/share/ocrd-resources/
- #- ./workflows:/app/workflows
- # - ./workflows/ocrd_workflows:/app/workflows/ocrd_workflows
+ #- ./models/ocrd-tesserocr-recognize:/usr/local/share/tessdata/
+ #- ./models/ocrd-calamari-recognize:/usr/local/share/ocrd-resources/
# - TODO/custom/data
\ No newline at end of file
diff --git a/prepare.sh b/prepare.sh
deleted file mode 100644
index cd60dcf..0000000
--- a/prepare.sh
+++ /dev/null
@@ -1,41 +0,0 @@
-#!/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 "$INNER_ZIP"
-
- echo "Done."
- done
-done
-
-echo " … and ready to go!"
diff --git a/scripts/convert-yml-to-json.py b/scripts/convert-yml-to-json.py
new file mode 100644
index 0000000..0ff8023
--- /dev/null
+++ b/scripts/convert-yml-to-json.py
@@ -0,0 +1,22 @@
+#!env python3
+
+# This script has been copied from https://github.com/OCR-D/spec/blob/master/scripts/yaml-to-json.py
+
+from yaml import safe_load
+from json import dumps
+from click import command, argument, option
+
+@command()
+@option('--indent', default=2, type=int)
+@argument('src')
+@argument('dst')
+def cli(src, dst, indent):
+ kwargs = {}
+ if indent > 0:
+ kwargs['indent'] = indent
+ with open(src, 'r', encoding='utf-8') as f_in, open(dst, 'w', encoding='utf-8') as f_out:
+ ret = safe_load(f_in)
+ f_out.write(dumps(ret, **kwargs))
+
+if __name__ == '__main__':
+ cli()
\ No newline at end of file
diff --git a/scripts/prepare.sh b/scripts/prepare.sh
new file mode 100644
index 0000000..287c1e5
--- /dev/null
+++ b/scripts/prepare.sh
@@ -0,0 +1,98 @@
+#!/bin/bash
+
+mkdir gt
+
+echo "Prepare OCR-D Ground Truth …"
+
+while IFS= read -r URL; do
+ OWNER=$(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 < data_srcs/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"
+ if [[ ! -d $NAME && $NAME != "reichsanzeiger-gt" ]]; then
+ unzip -qq -d "$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 "Recreate required directory structure for $INNER_ZIP_NAME."
+ mkdir "$NAME"/"$INNER_ZIP_NAME"/data/"$INNER_ZIP_NAME"
+ mv "$NAME"/"$INNER_ZIP_NAME"/data/OCR-* "$NAME"/"$INNER_ZIP_NAME"/data/"$INNER_ZIP_NAME"
+ mv "$NAME"/"$INNER_ZIP_NAME"/data/mets.xml "$NAME"/"$INNER_ZIP_NAME"/data/"$INNER_ZIP_NAME"
+ cp "$NAME"/metadata.json "$NAME"/"$INNER_ZIP_NAME"/data/"$INNER_ZIP_NAME"/metadata.json
+
+ echo "Moving $INNER_ZIP_NAME higher in dir structure."
+ mv "$NAME"/"$INNER_ZIP_NAME" .
+ echo "Done."
+ done
+ rm -rf "$NAME"
+ fi
+done
+
+echo "Prepare Reichsanzeiger GT …"
+
+if [[ $1 == "ra-full" ]]; then
+ echo "Preparing the full Reichsanzeiger GT."
+
+ if [ ! -d reichsanzeiger-gt ]; then
+ git clone https://github.com/UB-Mannheim/reichsanzeiger-gt
+ fi
+
+ RA_GT=/app/gt/reichsanzeiger-gt
+ DATA_DIR=/$RA_GT/data
+ cd $DATA_DIR|| exit
+
+ if [[ -d reichsanzeiger-1820-1939/OCR-D-IMG ]]; then
+ echo "Skip downloading Reichsanzeiger images."
+ else
+ bash download_images.sh
+ fi
+ cd reichsanzeiger-1820-1939 || exit
+ ocrd workspace init
+ mkdir OCR-D-IMG
+ cp ../images/* OCR-D-IMG
+ rm -rf ../images
+ rm -rf ../reichsanzeiger-1820-1939_with-TableRegion
+ cp -r GT-PAGE OCR-D-GT-SEG-LINE
+
+ echo "Adding images to mets …"
+
+ FILEGRP="OCR-D-IMG"
+ EXT=".jpg" # the actual extension of the image files
+ MEDIATYPE='image/jpeg' # the actual media type of the image files
+ for i in "$FILEGRP"/*"$EXT"; do
+ BASE=$(basename "${i}" $EXT);
+ ocrd workspace add -G $FILEGRP -i ${FILEGRP}_"${BASE}" -g P_"${BASE}" -m $MEDIATYPE "${i}";
+ done
+
+ python3 /app/scripts/convert-yml-to-json.py --indent 2 $RA_GT/METADATA.yml $RA_GT/metadata.json
+
+ echo " … and ready to go!"
+
+else
+ echo "Prepare smaller sets of Reichsanzeiger GT."
+ cd /app || exit
+ bash /app/scripts/prepare_reichsanzeiger_sets.sh
+fi
diff --git a/scripts/prepare_reichsanzeiger_sets.sh b/scripts/prepare_reichsanzeiger_sets.sh
new file mode 100644
index 0000000..a5667f4
--- /dev/null
+++ b/scripts/prepare_reichsanzeiger_sets.sh
@@ -0,0 +1,68 @@
+#!/bin/sh
+
+echo "Download Reichsanzeiger GT repository."
+cd gt || exit
+if [ ! -d reichsanzeiger-gt ]; then
+ git clone https://github.com/UB-Mannheim/reichsanzeiger-gt
+fi
+
+cd .. || exit
+
+
+PREFIX="data_srcs"
+files=(
+ "$PREFIX"/reichsanzeiger_many_ads.list
+ "$PREFIX"/reichsanzeiger_random.list
+ "$PREFIX"/reichsanzeiger_tables.list
+ "$PREFIX"/reichsanzeiger_title_pages.list
+)
+
+for FILE in "${files[@]}"; do
+ NAME=$(basename "$FILE" .list)
+ if [ -d gt/"$NAME" ]; then
+ echo "Directory gt/$NAME already exists. Skipping download."
+ else
+ echo "Processing $FILE."
+ mkdir -p gt/"$NAME"/data/"$NAME"/OCR-D-IMG
+ mkdir -p gt/"$NAME"/data/"$NAME"/OCR-D-GT-SEG-LINE
+
+ urlbase=$(echo "aHR0cHM6Ly9kaWdpLmJpYi51bmktbWFubmhlaW0uZGUvcmVpY2hzYW56ZWlnZXIuZmNnaT9GSUY9
+L3JlaWNoc2FuemVpZ2VyL2ZpbG0vCg==" | base64 -d)
+ while read -r line; do
+ wget --limit-rate=500k "${urlbase}${line% *}" -O ./gt/"$NAME"/data/"$NAME"/OCR-D-IMG/"${line#* }"
+ IMG_NAME=$(basename "${line#* }" .jpg)
+ cp gt/reichsanzeiger-gt/data/reichsanzeiger-1820-1939/GT-PAGE/"$IMG_NAME".xml gt/"$NAME"/data/"$NAME"/OCR-D-GT-SEG-LINE/"$IMG_NAME".xml
+ done < "$FILE"
+ fi
+
+ if [ ! -f gt/"$NAME"/mets.xml ]; then
+ echo "Preparing OCR-D workspace for $NAME".
+ cd gt/"$NAME"/data/"$NAME" || exit
+ ocrd workspace init
+ ocrd workspace set-id "$NAME"
+
+ FILEGRP="OCR-D-IMG"
+ FILEGRP_2="OCR-D-GT-SEG-LINE"
+ # add images to mets
+ EXT=".jpg" # the actual extension of the image files
+ MEDIATYPE='image/jpeg' # the actual media type of the image files
+ for i in OCR-D-IMG/*"$EXT"; do
+ BASE=$(basename "${i}" $EXT)
+ ocrd workspace add -G $FILEGRP -i "${FILEGRP}"_"${BASE}" -g P_"${BASE}" -m $MEDIATYPE "${i}"
+ done
+
+ # add GT to mets
+ for i in "$FILEGRP_2"/*.xml; do
+ BASE=$(basename "${i}" ".xml")
+ ocrd workspace add -G $FILEGRP_2 -i "${FILEGRP_2}"_"${BASE}" -g P_"${BASE}" -m text/xml "${i}"
+ done
+ fi
+
+ if [ ! -f gt/"$NAME"/data/"$NAME"/metadata.json ]; then
+ cp /app/gt/reichsanzeiger-gt/METADATA.yml /app/gt/"$NAME"/data/"$NAME"/METADATA.yml
+ python3 /app/scripts/convert-yml-to-json.py --indent 2 /app/gt/"$NAME"/data/"$NAME"/METADATA.yml /app/gt/"$NAME"/data/"$NAME"/metadata.json
+ fi
+ cd /app || exit
+done
+
+echo "Preparation of Reichsanzeiger GT subsets done."
diff --git a/src/benchmark_extraction.py b/src/benchmark_extraction.py
index 2dfe6c2..39b4b21 100644
--- a/src/benchmark_extraction.py
+++ b/src/benchmark_extraction.py
@@ -106,17 +106,8 @@ def get_eval_tool(mets_path: str) -> str:
def get_gt_workspace(workspace_path: str) -> Dict[str, str]:
current_workspace = get_workspace_name(workspace_path)
- split_workspace_name = current_workspace.split('_')
- workspace_name_wo_workflow = split_workspace_name[0] + '_' + split_workspace_name[1] + '_' + split_workspace_name[2]
- font = ''
- if split_workspace_name[1] == 'ant':
- font = 'Antiqua'
- elif split_workspace_name[1] == 'frak':
- font = 'Fraktur'
- else:
- font = 'Font Mix'
- url = 'https://github.com/OCR-D/quiver-data/blob/main/' + workspace_name_wo_workflow + '.ocrd.zip'
- label = f'GT workspace {split_workspace_name[0]}th century {font} {split_workspace_name[2]} layout'
+ url = 'https://github.com/OCR-D/quiver-data/blob/main/' + current_workspace + '.ocrd.zip'
+ label = f'GT workspace {current_workspace}'
return {
'@id': url,
'label': label
@@ -196,7 +187,8 @@ def get_nextflow_completed_process_file(workspace_path: str):
return file
def get_nextflow_time(workspace_path: str, time_type: str) -> float:
- files = listdir(workspace_path)
+ highest_workspace_dir = '/'.join(workspace_path.split('/')[:-2])
+ files = listdir(highest_workspace_dir)
logs = []
for file in files:
if '.command.log' in file:
@@ -204,7 +196,7 @@ def get_nextflow_time(workspace_path: str, time_type: str) -> float:
time_per_workflow_step = []
for log in logs:
- with open(workspace_path + '/' + log, 'r', encoding='utf-8') as l:
+ with open(highest_workspace_dir + '/' + log, 'r', encoding='utf-8') as l:
log_file = l.read()
no_sec_s = re.search(rf'([0-9]+?\.[0-9]+?)s \({time_type}\)', log_file).group(1)
time_per_workflow_step.append(float(no_sec_s))
diff --git a/workflows/execute_workflows.sh b/workflows/execute_workflows.sh
index 5f3f59f..db8957d 100755
--- a/workflows/execute_workflows.sh
+++ b/workflows/execute_workflows.sh
@@ -39,9 +39,9 @@ convert_ocrd_wfs_to_NextFlow() {
download_models() {
echo "Download the necessary models if not available"
- if [[ ! -f /usr/local/share/ocrd-resources/ocrd-tesserocr-recognize/ ]]
+ if [[ ! -d /usr/local/share/tessdata ]]
then
- mkdir -p /usr/local/share/ocrd-resources/
+ #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 ]]
@@ -58,32 +58,15 @@ create_wf_specific_workspaces() {
# 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
+ if [[ ! $DIR_NAME == "reichsanzeiger-gt" ]]; then
+ echo "Create workflow specific workspace for $DIR_NAME."
for WORKFLOW in "$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"_"$WF_NAME"/
+ cp "$WORKFLOW" "$WORKSPACE_DIR"/tmp/"$DIR_NAME"_"$WF_NAME"/data/*/
done
fi
done
@@ -93,11 +76,10 @@ clean_up_tmp_dirs() {
echo "Clean up intermediate dirs …"
for DIR in "$WORKSPACE_DIR"/tmp/*
do
+ echo "Cleaning up $DIR."
DIR_NAME=$(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"/"$DIR_NAME"/
+ cp "$OCRD_WORKFLOW_DIR"/*eval.txt.nf "$WORKSPACE_DIR"/"$DIR_NAME"/data/*/
done
rm -rf "$WORKSPACE_DIR"/tmp
@@ -114,16 +96,16 @@ execute_wfs_and_extract_benchmarks() {
DIR_NAME=$(basename $WS_DIR)
- run "$WS_DIR"/*ocr.txt.nf "$DIR_NAME" "$WS_DIR"
- run "$WS_DIR"/*eval.txt.nf "$DIR_NAME" "$WS_DIR"
+ run "$WS_DIR"/data/*/*ocr.txt.nf "$DIR_NAME" "$WS_DIR"
+ run "$WS_DIR"/data/*/*eval.txt.nf "$DIR_NAME" "$WS_DIR"
# create a result JSON according to the specs
echo "Get Benchmark JSON …"
- quiver benchmarks-extraction "$WS_DIR" "$WORKFLOW"
+ quiver benchmarks-extraction "$WS_DIR"/data/* "$WORKFLOW"
echo "Done."
# move data to results dir
- mv "$WS_DIR"/*.json "$WORKFLOW_DIR"/results
+ mv "$WS_DIR"/data/*/*.json "$WORKFLOW_DIR"/results
fi
done
cd "$ROOT" || exit
@@ -132,7 +114,7 @@ execute_wfs_and_extract_benchmarks() {
adjust_workflow_settings() {
# $1: $WORKFLOW
# $2: $DIR_NAME
- sed -i "s CURRENT app/workflows/workspaces/$2 g" "$1"
+ sed -i "s CURRENT app/workflows/workspaces/$2/data/*/ g" "$1"
}
rename_and_move_nextflow_result() {
@@ -142,13 +124,13 @@ rename_and_move_nextflow_result() {
WORKFLOW_NAME=$(basename -s .txt.nf "$1")
rm "$WORKFLOW_DIR"/nf-results/*process_completed.json
mv "$WORKFLOW_DIR"/nf-results/*_completed.json "$WORKFLOW_DIR"/results/"$2"_"$WORKFLOW_NAME"_completed.json
- if [ $WORKFLOW_NAME != "dinglehopper_eval" ]; then
+ if [ "$WORKFLOW_NAME" != "dinglehopper_eval" ]; then
for DIR in "$WORKSPACE_DIR"/work/*
do
- WORK_DIR_NAME=$(basename $DIR)
- for SUB_WORK_DIR in $DIR/*
+ WORK_DIR_NAME=$(basename "$DIR")
+ for SUB_WORK_DIR in "$DIR"/*
do
- SUB_WORK_DIR_NAME=$(basename $SUB_WORK_DIR)
+ SUB_WORK_DIR_NAME=$(basename "$SUB_WORK_DIR")
mv "$WORKSPACE_DIR"/work/"$WORK_DIR_NAME"/"$SUB_WORK_DIR_NAME"/.command.log "$WORKSPACE_DIR"/"$2"/"$WORK_DIR_NAME"_"$SUB_WORK_DIR_NAME".command.log
done
@@ -165,7 +147,7 @@ run() {
adjust_workflow_settings "$1" "$2"
nextflow run "$1" -with-weblog http://127.0.0.1:8000/nextflow/
rename_and_move_nextflow_result "$1" "$2"
- save_workspaces "$3" "$2" "$1"
+ save_workspaces "$3"/data "$2" "$1"
}
save_workspaces() {
@@ -173,7 +155,7 @@ save_workspaces() {
# $2: $DIR_NAME
# $3: $WORKFLOW
echo "Zipping workspace $1"
- ocrd zip bag -d $1 -i $1 $1
+ ocrd zip bag -d "$DIR_NAME"/data/* -i "$DIR_NAME"/data/* "$DIR_NAME"
WORKFLOW_NAME=$(basename -s .txt.nf "$3")
mv "$WORKSPACE_DIR"/"$2".zip "$WORKFLOW_DIR"/results/"$2"_"$WORKFLOW_NAME".zip
}
--
GitLab