Machine based reading order integration

.github/workflows/test-eynollah.yml

@@ -14,6 +14,12 @@ jobs:
         python-version: ['3.8', '3.9', '3.10', '3.11']
 
     steps:
+    - name: clean up
+      run: |
+        sudo rm -rf /usr/share/dotnet
+        sudo rm -rf /opt/ghc
+        sudo rm -rf "/usr/local/share/boost"
+        sudo rm -rf "$AGENT_TOOLSDIRECTORY"
     - uses: actions/checkout@v4
     - uses: actions/cache@v4
       id: model_cache

Makefile

@@ -32,9 +32,9 @@ models_eynollah: models_eynollah.tar.gz
 models_eynollah.tar.gz:
 	# wget 'https://qurator-data.de/eynollah/2021-04-25/models_eynollah.tar.gz'
 	# wget 'https://qurator-data.de/eynollah/2022-04-05/models_eynollah_renamed.tar.gz'
-	# wget 'https://qurator-data.de/eynollah/2022-04-05/models_eynollah_renamed_savedmodel.tar.gz'
+	wget 'https://qurator-data.de/eynollah/2022-04-05/models_eynollah.tar.gz'
 	# wget 'https://github.com/qurator-spk/eynollah/releases/download/v0.3.0/models_eynollah.tar.gz'
-	wget 'https://github.com/qurator-spk/eynollah/releases/download/v0.3.1/models_eynollah.tar.gz'
+	# wget 'https://github.com/qurator-spk/eynollah/releases/download/v0.3.1/models_eynollah.tar.gz'
 
 # Install with pip
 install:
@@ -45,7 +45,7 @@ install-dev:
 	pip install -e .
 
 smoke-test:
-	eynollah -i tests/resources/kant_aufklaerung_1784_0020.tif -o . -m $(PWD)/models_eynollah
+	eynollah layout -i tests/resources/kant_aufklaerung_1784_0020.tif -o . -m $(PWD)/models_eynollah
 
 # Run unit tests
 test:

pyproject.toml

@@ -28,6 +28,7 @@ classifiers = [
 [project.scripts]
 eynollah = "eynollah.cli:main"
 ocrd-eynollah-segment = "eynollah.ocrd_cli:main"
+ocrd-sbb-binarize = "eynollah.ocrd_cli_binarization:cli"
 
 [project.urls]
 Homepage = "https://github.com/qurator-spk/eynollah"

requirements.txt

@@ -2,7 +2,10 @@
 ocrd >= 2.23.3
 numpy <1.24.0
 scikit-learn >= 0.23.2
-tensorflow == 2.12.1
+tensorflow < 2.13
 imutils >= 0.5.3
 matplotlib
 setuptools >= 50
+transformers <= 4.30.2
+torch <= 2.0.1
+numba <= 0.58.1

src/eynollah/cli.py

@@ -2,15 +2,94 @@
 import click
 from ocrd_utils import initLogging, setOverrideLogLevel
 from eynollah.eynollah import Eynollah
+from eynollah.sbb_binarize import SbbBinarizer
 
+@click.group()
+def main():
+    pass
 
-@click.command()
+@main.command()
+@click.option(
+    "--dir_xml",
+    "-dx",
+    help="directory of GT page-xml files",
+    type=click.Path(exists=True, file_okay=False),
+)
+
+@click.option(
+    "--dir_out_modal_image",
+    "-domi",
+    help="directory where ground truth images would be written",
+    type=click.Path(exists=True, file_okay=False),
+)
+
+@click.option(
+    "--dir_out_classes",
+    "-docl",
+    help="directory where ground truth classes would be written",
+    type=click.Path(exists=True, file_okay=False),
+)
+
+@click.option(
+    "--input_height",
+    "-ih",
+    help="input height",
+)
+@click.option(
+    "--input_width",
+    "-iw",
+    help="input width",
+)
+@click.option(
+    "--min_area_size",
+    "-min",
+    help="min area size of regions considered for reading order training.",
+)
+
+def machine_based_reading_order(dir_xml, dir_out_modal_image, dir_out_classes, input_height, input_width, min_area_size):
+    xml_files_ind = os.listdir(dir_xml)
+
+@main.command()
+@click.option('--patches/--no-patches', default=True, help='by enabling this parameter you let the model to see the image in patches.')
+
+@click.option('--model_dir', '-m', type=click.Path(exists=True, file_okay=False), required=True, help='directory containing models for prediction')
+
+@click.argument('input_image')
+
+@click.argument('output_image')
+@click.option(
+    "--dir_in",
+    "-di",
+    help="directory of images",
+    type=click.Path(exists=True, file_okay=False),
+)
+@click.option(
+    "--dir_out",
+    "-do",
+    help="directory where the binarized images will be written",
+    type=click.Path(exists=True, file_okay=False),
+)
+
+def binarization(patches, model_dir, input_image, output_image, dir_in, dir_out):
+    if not dir_out and (dir_in):
+        print("Error: You used -di but did not set -do")
+        sys.exit(1)
+    elif dir_out and not (dir_in):
+        print("Error: You used -do to write out binarized images but have not set -di")
+        sys.exit(1)
+    SbbBinarizer(model_dir).run(image_path=input_image, use_patches=patches, save=output_image, dir_in=dir_in, dir_out=dir_out)
+
+
+
+
+@main.command()
 @click.option(
     "--image",
     "-i",
     help="image filename",
     type=click.Path(exists=True, dir_okay=False),
 )
+
 @click.option(
     "--out",
     "-o",
@@ -140,36 +219,41 @@
     help="if this parameter set to true, this tool would ignore page extraction",
 )
 @click.option(
-    "--log-level",
+    "--reading_order_machine_based/--heuristic_reading_order",
+    "-romb/-hro",
+    is_flag=True,
+    help="if this parameter set to true, this tool would apply machine based reading order detection",
+)
+@click.option(
+    "--do_ocr",
+    "-ocr/-noocr",
+    is_flag=True,
+    help="if this parameter set to true, this tool will try to do ocr",
+)
+@click.option(
+    "--num_col_upper",
+    "-ncu",
+    help="lower limit of columns in document image",
+)
+@click.option(
+    "--num_col_lower",
+    "-ncl",
+    help="upper limit of columns in document image",
+)
+@click.option(
+    "--skip_layout_and_reading_order",
+    "-slro/-noslro",
+    is_flag=True,
+    help="if this parameter set to true, this tool will ignore layout detection and reading order. It means that textline detection will be done within printspace and contours of textline will be written in xml output file.",
+)
+@click.option(
+    "--log_level",
     "-l",
     type=click.Choice(['OFF', 'DEBUG', 'INFO', 'WARN', 'ERROR']),
     help="Override log level globally to this",
 )
-def main(
-    image,
-    out,
-    dir_in,
-    model,
-    save_images,
-    save_layout,
-    save_deskewed,
-    save_all,
-    extract_only_images,
-    save_page,
-    enable_plotting,
-    allow_enhancement,
-    curved_line,
-    textline_light,
-    full_layout,
-    tables,
-    right2left,
-    input_binary,
-    allow_scaling,
-    headers_off,
-    light_version,
-    ignore_page_extraction,
-    log_level
-):
+
+def layout(image, out, dir_in, model, save_images, save_layout, save_deskewed, save_all, extract_only_images, save_page, enable_plotting, allow_enhancement, curved_line, textline_light, full_layout, tables, right2left, input_binary, allow_scaling, headers_off, light_version, reading_order_machine_based, do_ocr, num_col_upper, num_col_lower, skip_layout_and_reading_order, ignore_page_extraction, log_level):
     if log_level:
         setOverrideLogLevel(log_level)
     initLogging()
@@ -182,6 +266,8 @@ def main(
     if textline_light and not light_version:
         print('Error: You used -tll to enable light textline detection but -light is not enabled')
         sys.exit(1)
+    if light_version and not textline_light:
+        print('Error: You used -light without -tll. Light version need light textline to be enabled.')
     if extract_only_images and  (allow_enhancement or allow_scaling or light_version or curved_line or textline_light or full_layout or tables or right2left or headers_off) :
         print('Error: You used -eoi which can not be enabled alongside light_version -light or allow_scaling -as or allow_enhancement -ae or curved_line -cl or textline_light -tll or full_layout -fl or tables -tab or right2left -r2l or headers_off -ho')
         sys.exit(1)
@@ -208,6 +294,11 @@ def main(
         headers_off=headers_off,
         light_version=light_version,
         ignore_page_extraction=ignore_page_extraction,
+        reading_order_machine_based=reading_order_machine_based,
+        do_ocr=do_ocr,
+        num_col_upper=num_col_upper,
+        num_col_lower=num_col_lower,
+        skip_layout_and_reading_order=skip_layout_and_reading_order,
     )
     if dir_in:
         eynollah.run()