mirror of
https://github.com/qurator-spk/eynollah.git
synced 2025-10-06 14:39:55 +02:00
Merge remote-tracking branch 'bertsky/mbro_dead_code-plus-fixes-plus-tests' into prepare-release-v0.5.0
This commit is contained in:
kba
commit
5e15c4f248
11 changed files with 4190 additions and 273 deletions
10
Makefile
10
Makefile
|
|
@ -82,13 +82,21 @@ smoke-test: tests/resources/kant_aufklaerung_1784_0020.tif
|
|||
eynollah layout -i $< -o $(TMPDIR) -m $(CURDIR)/models_eynollah
|
||||
fgrep -q http://schema.primaresearch.org/PAGE/gts/pagecontent/2019-07-15 $(TMPDIR)/$(basename $(<F)).xml
|
||||
fgrep -c -e TextRegion -e ImageRegion -e SeparatorRegion $(TMPDIR)/$(basename $(<F)).xml
|
||||
# directory mode (skip one, add one):
|
||||
# layout, directory mode (skip one, add one):
|
||||
eynollah layout -di $(<D) -o $(TMPDIR) -m $(CURDIR)/models_eynollah
|
||||
test -s $(TMPDIR)/euler_rechenkunst01_1738_0025.xml
|
||||
# mbreorder, directory mode (overwrite):
|
||||
eynollah machine-based-reading-order -di $(<D) -o $(TMPDIR) -m $(CURDIR)/models_eynollah
|
||||
fgrep -q http://schema.primaresearch.org/PAGE/gts/pagecontent/2019-07-15 $(TMPDIR)/$(basename $(<F)).xml
|
||||
fgrep -c -e RegionRefIndexed $(TMPDIR)/$(basename $(<F)).xml
|
||||
# binarize:
|
||||
eynollah binarization -m $(CURDIR)/default-2021-03-09 -i $< -o $(TMPDIR)/$(<F)
|
||||
test -s $(TMPDIR)/$(<F)
|
||||
@set -x; test "$$(identify -format '%w %h' $<)" = "$$(identify -format '%w %h' $(TMPDIR)/$(<F))"
|
||||
# enhance:
|
||||
eynollah enhancement -m $(CURDIR)/models_eynollah -sos -i $< -o $(TMPDIR) -O
|
||||
test -s $(TMPDIR)/$(<F)
|
||||
@set -x; test "$$(identify -format '%w %h' $<)" = "$$(identify -format '%w %h' $(TMPDIR)/$(<F))"
|
||||
$(RM) -r $(TMPDIR)
|
||||
|
||||
ocrd-test: export OCRD_MISSING_OUTPUT := ABORT
|
||||
|
|
|
|||
|
|
@ -13,22 +13,23 @@ def main():
|
|||
|
||||
@main.command()
|
||||
@click.option(
|
||||
"--dir_xml",
|
||||
"-dx",
|
||||
help="directory of page-xml files",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
"--xml_file",
|
||||
"-xml",
|
||||
help="xml filename",
|
||||
"--input",
|
||||
"-i",
|
||||
help="PAGE-XML input filename",
|
||||
type=click.Path(exists=True, dir_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
"--dir_out",
|
||||
"-do",
|
||||
"--dir_in",
|
||||
"-di",
|
||||
help="directory of PAGE-XML input files (instead of --input)",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
"--out",
|
||||
"-o",
|
||||
help="directory for output images",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
required=True,
|
||||
)
|
||||
@click.option(
|
||||
"--model",
|
||||
|
|
@ -37,24 +38,38 @@ def main():
|
|||
type=click.Path(exists=True, file_okay=False),
|
||||
required=True,
|
||||
)
|
||||
@click.option(
|
||||
"--log_level",
|
||||
"-l",
|
||||
type=click.Choice(['OFF', 'DEBUG', 'INFO', 'WARN', 'ERROR']),
|
||||
help="Override log level globally to this",
|
||||
)
|
||||
|
||||
def machine_based_reading_order(dir_xml, xml_file, dir_out, model):
|
||||
raedingorder_object = machine_based_reading_order_on_layout(model, dir_out=dir_out, logger=getLogger('enhancement'))
|
||||
def machine_based_reading_order(input, dir_in, out, model, log_level):
|
||||
assert bool(input) != bool(dir_in), "Either -i (single input) or -di (directory) must be provided, but not both."
|
||||
orderer = machine_based_reading_order_on_layout(model, dir_out=out)
|
||||
if log_level:
|
||||
orderer.logger.setLevel(getLevelName(log_level))
|
||||
|
||||
if dir_xml:
|
||||
raedingorder_object.run(dir_in=dir_xml)
|
||||
if dir_in:
|
||||
orderer.run(dir_in=dir_in)
|
||||
else:
|
||||
raedingorder_object.run(xml_filename=xml_file)
|
||||
orderer.run(xml_filename=input)
|
||||
|
||||
|
||||
@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.option("--input-image", "-i", help="input image", type=click.Path(exists=True, dir_okay=False))
|
||||
@click.option(
|
||||
"--input-image", "--image",
|
||||
"-i",
|
||||
help="input image filename",
|
||||
type=click.Path(exists=True, dir_okay=False)
|
||||
)
|
||||
@click.option(
|
||||
"--dir_in",
|
||||
"-di",
|
||||
help="directory of input images",
|
||||
help="directory of input images (instead of --image)",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
|
|
@ -62,25 +77,34 @@ def machine_based_reading_order(dir_xml, xml_file, dir_out, model):
|
|||
"-o",
|
||||
help="output image (if using -i) or output image directory (if using -di)",
|
||||
type=click.Path(file_okay=True, dir_okay=True),
|
||||
required=True,
|
||||
)
|
||||
def binarization(patches, model_dir, input_image, dir_in, output):
|
||||
assert (dir_in is None) != (input_image is None), "Specify either -di and or -i not both"
|
||||
SbbBinarizer(model_dir).run(image_path=input_image, use_patches=patches, output=output, dir_in=dir_in)
|
||||
|
||||
@click.option(
|
||||
"--log_level",
|
||||
"-l",
|
||||
type=click.Choice(['OFF', 'DEBUG', 'INFO', 'WARN', 'ERROR']),
|
||||
help="Override log level globally to this",
|
||||
)
|
||||
def binarization(patches, model_dir, input_image, dir_in, output, log_level):
|
||||
assert bool(input_image) != bool(dir_in), "Either -i (single input) or -di (directory) must be provided, but not both."
|
||||
binarizer = SbbBinarizer(model_dir)
|
||||
if log_level:
|
||||
binarizer.log.setLevel(getLevelName(log_level))
|
||||
binarizer.run(image_path=input_image, use_patches=patches, output=output, dir_in=dir_in)
|
||||
|
||||
|
||||
@main.command()
|
||||
@click.option(
|
||||
"--image",
|
||||
"-i",
|
||||
help="image filename",
|
||||
help="input image filename",
|
||||
type=click.Path(exists=True, dir_okay=False),
|
||||
)
|
||||
|
||||
@click.option(
|
||||
"--out",
|
||||
"-o",
|
||||
help="directory to write output xml data",
|
||||
help="directory for output PAGE-XML files",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
required=True,
|
||||
)
|
||||
|
|
@ -93,7 +117,7 @@ def binarization(patches, model_dir, input_image, dir_in, output):
|
|||
@click.option(
|
||||
"--dir_in",
|
||||
"-di",
|
||||
help="directory of images",
|
||||
help="directory of input images (instead of --image)",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
|
|
@ -128,35 +152,34 @@ def binarization(patches, model_dir, input_image, dir_in, output):
|
|||
)
|
||||
|
||||
def enhancement(image, out, overwrite, dir_in, model, num_col_upper, num_col_lower, save_org_scale, log_level):
|
||||
assert bool(image) != bool(dir_in), "Either -i (single input) or -di (directory) must be provided, but not both."
|
||||
initLogging()
|
||||
if log_level:
|
||||
getLogger('enhancement').setLevel(getLevelName(log_level))
|
||||
assert image or dir_in, "Either a single image -i or a dir_in -di is required"
|
||||
enhancer_object = Enhancer(
|
||||
enhancer = Enhancer(
|
||||
model,
|
||||
logger=getLogger('enhancement'),
|
||||
dir_out=out,
|
||||
num_col_upper=num_col_upper,
|
||||
num_col_lower=num_col_lower,
|
||||
save_org_scale=save_org_scale,
|
||||
)
|
||||
if log_level:
|
||||
enhancer.logger.setLevel(getLevelName(log_level))
|
||||
if dir_in:
|
||||
enhancer_object.run(dir_in=dir_in, overwrite=overwrite)
|
||||
enhancer.run(dir_in=dir_in, overwrite=overwrite)
|
||||
else:
|
||||
enhancer_object.run(image_filename=image, overwrite=overwrite)
|
||||
enhancer.run(image_filename=image, overwrite=overwrite)
|
||||
|
||||
@main.command()
|
||||
@click.option(
|
||||
"--image",
|
||||
"-i",
|
||||
help="image filename",
|
||||
help="input image filename",
|
||||
type=click.Path(exists=True, dir_okay=False),
|
||||
)
|
||||
|
||||
@click.option(
|
||||
"--out",
|
||||
"-o",
|
||||
help="directory to write output xml data",
|
||||
help="directory for output PAGE-XML files",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
required=True,
|
||||
)
|
||||
|
|
@ -169,7 +192,7 @@ def enhancement(image, out, overwrite, dir_in, model, num_col_upper, num_col_low
|
|||
@click.option(
|
||||
"--dir_in",
|
||||
"-di",
|
||||
help="directory of images",
|
||||
help="directory of input images (instead of --image)",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
|
|
@ -360,8 +383,6 @@ def layout(image, out, overwrite, dir_in, model, save_images, save_layout, save_
|
|||
getLogger('eynollah').setLevel(logging.INFO)
|
||||
else:
|
||||
initLogging()
|
||||
if log_level:
|
||||
getLogger('eynollah').setLevel(getLevelName(log_level))
|
||||
assert enable_plotting or not save_layout, "Plotting with -sl also requires -ep"
|
||||
assert enable_plotting or not save_deskewed, "Plotting with -sd also requires -ep"
|
||||
assert enable_plotting or not save_all, "Plotting with -sa also requires -ep"
|
||||
|
|
@ -380,16 +401,10 @@ def layout(image, out, overwrite, dir_in, model, save_images, save_layout, save_
|
|||
assert not extract_only_images or not tables, "Image extraction -eoi can not be set alongside tables -tab"
|
||||
assert not extract_only_images or not right2left, "Image extraction -eoi can not be set alongside right2left -r2l"
|
||||
assert not extract_only_images or not headers_off, "Image extraction -eoi can not be set alongside headers_off -ho"
|
||||
assert image or dir_in, "Either a single image -i or a dir_in -di is required"
|
||||
assert bool(image) != bool(dir_in), "Either -i (single input) or -di (directory) must be provided, but not both."
|
||||
eynollah = Eynollah(
|
||||
model,
|
||||
dir_out=out,
|
||||
dir_of_cropped_images=save_images,
|
||||
extract_only_images=extract_only_images,
|
||||
dir_of_layout=save_layout,
|
||||
dir_of_deskewed=save_deskewed,
|
||||
dir_of_all=save_all,
|
||||
dir_save_page=save_page,
|
||||
enable_plotting=enable_plotting,
|
||||
allow_enhancement=allow_enhancement,
|
||||
curved_line=curved_line,
|
||||
|
|
@ -412,56 +427,64 @@ def layout(image, out, overwrite, dir_in, model, save_images, save_layout, save_
|
|||
threshold_art_class_textline=threshold_art_class_textline,
|
||||
threshold_art_class_layout=threshold_art_class_layout,
|
||||
)
|
||||
if dir_in:
|
||||
eynollah.run(dir_in=dir_in, overwrite=overwrite)
|
||||
else:
|
||||
eynollah.run(image_filename=image, overwrite=overwrite)
|
||||
|
||||
if log_level:
|
||||
eynollah.logger.setLevel(getLevelName(log_level))
|
||||
eynollah.run(overwrite=overwrite,
|
||||
image_filename=image,
|
||||
dir_in=dir_in,
|
||||
dir_out=out,
|
||||
dir_of_cropped_images=save_images,
|
||||
dir_of_layout=save_layout,
|
||||
dir_of_deskewed=save_deskewed,
|
||||
dir_of_all=save_all,
|
||||
dir_save_page=save_page,
|
||||
)
|
||||
|
||||
@main.command()
|
||||
@click.option(
|
||||
"--image",
|
||||
"-i",
|
||||
help="image filename",
|
||||
help="input image filename",
|
||||
type=click.Path(exists=True, dir_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
"--dir_in",
|
||||
"-di",
|
||||
help="directory of input images (instead of --image)",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
"--dir_in_bin",
|
||||
"-dib",
|
||||
help="directory of binarized images (in addition to --dir_in for RGB images; filename stems must match the RGB image files, with '.png' suffix).\nPerform prediction using both RGB and binary images. (This does not necessarily improve results, however it may be beneficial for certain document images.)",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
"--dir_xmls",
|
||||
"-dx",
|
||||
help="directory of input PAGE-XML files (in addition to --dir_in; filename stems must match the image files, with '.xml' suffix).",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
required=True,
|
||||
)
|
||||
@click.option(
|
||||
"--out",
|
||||
"-o",
|
||||
help="directory for output PAGE-XML files",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
required=True,
|
||||
)
|
||||
@click.option(
|
||||
"--dir_out_image_text",
|
||||
"-doit",
|
||||
help="directory for output images, newly rendered with predicted text",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
"--overwrite",
|
||||
"-O",
|
||||
help="overwrite (instead of skipping) if output xml exists",
|
||||
is_flag=True,
|
||||
)
|
||||
@click.option(
|
||||
"--dir_in",
|
||||
"-di",
|
||||
help="directory of images",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
"--dir_in_bin",
|
||||
"-dib",
|
||||
help="directory of binarized images. This should be given if you want to do prediction based on both rgb and bin images. And all bin images are png files",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
"--out",
|
||||
"-o",
|
||||
help="directory to write output xml data",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
required=True,
|
||||
)
|
||||
@click.option(
|
||||
"--dir_xmls",
|
||||
"-dx",
|
||||
help="directory of xmls",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
"--dir_out_image_text",
|
||||
"-doit",
|
||||
help="directory of images with predicted text",
|
||||
type=click.Path(exists=True, file_okay=False),
|
||||
)
|
||||
@click.option(
|
||||
"--model",
|
||||
"-m",
|
||||
|
|
@ -491,12 +514,6 @@ def layout(image, out, overwrite, dir_in, model, save_images, save_layout, save_
|
|||
is_flag=True,
|
||||
help="if this parameter set to true, cropped textline images will not be masked with textline contour.",
|
||||
)
|
||||
@click.option(
|
||||
"--prediction_with_both_of_rgb_and_bin",
|
||||
"-brb/-nbrb",
|
||||
is_flag=True,
|
||||
help="If this parameter is set to True, the prediction will be performed using both RGB and binary images. However, this does not necessarily improve results; it may be beneficial for certain document images.",
|
||||
)
|
||||
@click.option(
|
||||
"--batch_size",
|
||||
"-bs",
|
||||
|
|
@ -519,37 +536,36 @@ def layout(image, out, overwrite, dir_in, model, save_images, save_layout, save_
|
|||
help="Override log level globally to this",
|
||||
)
|
||||
|
||||
def ocr(image, overwrite, dir_in, dir_in_bin, out, dir_xmls, dir_out_image_text, model, model_name, tr_ocr, export_textline_images_and_text, do_not_mask_with_textline_contour, prediction_with_both_of_rgb_and_bin, batch_size, dataset_abbrevation, min_conf_value_of_textline_text, log_level):
|
||||
def ocr(image, dir_in, dir_in_bin, dir_xmls, out, dir_out_image_text, overwrite, model, model_name, tr_ocr, export_textline_images_and_text, do_not_mask_with_textline_contour, batch_size, dataset_abbrevation, min_conf_value_of_textline_text, log_level):
|
||||
initLogging()
|
||||
if log_level:
|
||||
getLogger('eynollah').setLevel(getLevelName(log_level))
|
||||
|
||||
assert not model or not model_name, "model directory -m can not be set alongside specific model name --model_name"
|
||||
assert bool(model) != bool(model_name), "Either -m (model directory) or --model_name (specific model name) must be provided."
|
||||
assert not export_textline_images_and_text or not tr_ocr, "Exporting textline and text -etit can not be set alongside transformer ocr -tr_ocr"
|
||||
assert not export_textline_images_and_text or not model, "Exporting textline and text -etit can not be set alongside model -m"
|
||||
assert not export_textline_images_and_text or not batch_size, "Exporting textline and text -etit can not be set alongside batch size -bs"
|
||||
assert not export_textline_images_and_text or not dir_in_bin, "Exporting textline and text -etit can not be set alongside directory of bin images -dib"
|
||||
assert not export_textline_images_and_text or not dir_out_image_text, "Exporting textline and text -etit can not be set alongside directory of images with predicted text -doit"
|
||||
assert not export_textline_images_and_text or not prediction_with_both_of_rgb_and_bin, "Exporting textline and text -etit can not be set alongside prediction with both rgb and bin -brb"
|
||||
assert (bool(image) ^ bool(dir_in)), "Either -i (single image) or -di (directory) must be provided, but not both."
|
||||
assert bool(image) != bool(dir_in), "Either -i (single image) or -di (directory) must be provided, but not both."
|
||||
eynollah_ocr = Eynollah_ocr(
|
||||
image_filename=image,
|
||||
dir_xmls=dir_xmls,
|
||||
dir_out_image_text=dir_out_image_text,
|
||||
dir_in=dir_in,
|
||||
dir_in_bin=dir_in_bin,
|
||||
dir_out=out,
|
||||
dir_models=model,
|
||||
model_name=model_name,
|
||||
tr_ocr=tr_ocr,
|
||||
export_textline_images_and_text=export_textline_images_and_text,
|
||||
do_not_mask_with_textline_contour=do_not_mask_with_textline_contour,
|
||||
prediction_with_both_of_rgb_and_bin=prediction_with_both_of_rgb_and_bin,
|
||||
batch_size=batch_size,
|
||||
pref_of_dataset=dataset_abbrevation,
|
||||
min_conf_value_of_textline_text=min_conf_value_of_textline_text,
|
||||
)
|
||||
eynollah_ocr.run(overwrite=overwrite)
|
||||
if log_level:
|
||||
eynollah_ocr.logger.setLevel(getLevelName(log_level))
|
||||
eynollah_ocr.run(overwrite=overwrite,
|
||||
dir_in=dir_in,
|
||||
dir_in_bin=dir_in_bin,
|
||||
image_filename=image,
|
||||
dir_xmls=dir_xmls,
|
||||
dir_out_image_text=dir_out_image_text,
|
||||
dir_out=out,
|
||||
)
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
|
|
|
|||
|
|
@ -107,6 +107,7 @@ from .utils.drop_capitals import (
|
|||
from .utils.marginals import get_marginals
|
||||
from .utils.resize import resize_image
|
||||
from .utils import (
|
||||
is_image_filename,
|
||||
boosting_headers_by_longshot_region_segmentation,
|
||||
crop_image_inside_box,
|
||||
find_num_col,
|
||||
|
|
@ -190,13 +191,7 @@ class Eynollah:
|
|||
def __init__(
|
||||
self,
|
||||
dir_models : str,
|
||||
dir_out : Optional[str] = None,
|
||||
dir_of_cropped_images : Optional[str] = None,
|
||||
extract_only_images : bool =False,
|
||||
dir_of_layout : Optional[str] = None,
|
||||
dir_of_deskewed : Optional[str] = None,
|
||||
dir_of_all : Optional[str] = None,
|
||||
dir_save_page : Optional[str] = None,
|
||||
enable_plotting : bool = False,
|
||||
allow_enhancement : bool = False,
|
||||
curved_line : bool = False,
|
||||
|
|
@ -220,18 +215,12 @@ class Eynollah:
|
|||
skip_layout_and_reading_order : bool = False,
|
||||
):
|
||||
self.logger = getLogger('eynollah')
|
||||
self.plotter = None
|
||||
|
||||
if skip_layout_and_reading_order:
|
||||
textline_light = True
|
||||
self.light_version = light_version
|
||||
self.dir_out = dir_out
|
||||
self.dir_of_all = dir_of_all
|
||||
self.dir_save_page = dir_save_page
|
||||
self.reading_order_machine_based = reading_order_machine_based
|
||||
self.dir_of_deskewed = dir_of_deskewed
|
||||
self.dir_of_deskewed = dir_of_deskewed
|
||||
self.dir_of_cropped_images=dir_of_cropped_images
|
||||
self.dir_of_layout=dir_of_layout
|
||||
self.enable_plotting = enable_plotting
|
||||
self.allow_enhancement = allow_enhancement
|
||||
self.curved_line = curved_line
|
||||
|
|
@ -422,21 +411,11 @@ class Eynollah:
|
|||
if dpi is not None:
|
||||
self.dpi = dpi
|
||||
|
||||
def reset_file_name_dir(self, image_filename):
|
||||
def reset_file_name_dir(self, image_filename, dir_out):
|
||||
t_c = time.time()
|
||||
self.cache_images(image_filename=image_filename)
|
||||
|
||||
self.plotter = None if not self.enable_plotting else EynollahPlotter(
|
||||
dir_out=self.dir_out,
|
||||
dir_of_all=self.dir_of_all,
|
||||
dir_save_page=self.dir_save_page,
|
||||
dir_of_deskewed=self.dir_of_deskewed,
|
||||
dir_of_cropped_images=self.dir_of_cropped_images,
|
||||
dir_of_layout=self.dir_of_layout,
|
||||
image_filename_stem=Path(Path(image_filename).name).stem)
|
||||
|
||||
self.writer = EynollahXmlWriter(
|
||||
dir_out=self.dir_out,
|
||||
dir_out=dir_out,
|
||||
image_filename=image_filename,
|
||||
curved_line=self.curved_line,
|
||||
textline_light = self.textline_light)
|
||||
|
|
@ -4602,7 +4581,17 @@ class Eynollah:
|
|||
return ordered_left_marginals, ordered_right_marginals, ordered_left_marginals_textline, ordered_right_marginals_textline, ordered_left_marginals_bbox, ordered_right_marginals_bbox, ordered_left_slopes_marginals, ordered_right_slopes_marginals
|
||||
|
||||
|
||||
def run(self, image_filename : Optional[str] = None, dir_in : Optional[str] = None, overwrite : bool = False):
|
||||
def run(self,
|
||||
overwrite: bool = False,
|
||||
image_filename: Optional[str] = None,
|
||||
dir_in: Optional[str] = None,
|
||||
dir_out: Optional[str] = None,
|
||||
dir_of_cropped_images: Optional[str] = None,
|
||||
dir_of_layout: Optional[str] = None,
|
||||
dir_of_deskewed: Optional[str] = None,
|
||||
dir_of_all: Optional[str] = None,
|
||||
dir_save_page: Optional[str] = None,
|
||||
):
|
||||
"""
|
||||
Get image and scales, then extract the page of scanned image
|
||||
"""
|
||||
|
|
@ -4623,20 +4612,37 @@ class Eynollah:
|
|||
enabled_modes.append("Table detection")
|
||||
if enabled_modes:
|
||||
self.logger.info("Enabled modes: " + ", ".join(enabled_modes))
|
||||
if self.enable_plotting:
|
||||
self.logger.info("Saving debug plots")
|
||||
if dir_of_cropped_images:
|
||||
self.logger.info(f"Saving cropped images to: {dir_of_cropped_images}")
|
||||
if dir_of_layout:
|
||||
self.logger.info(f"Saving layout plots to: {dir_of_layout}")
|
||||
if dir_of_deskewed:
|
||||
self.logger.info(f"Saving deskewed images to: {dir_of_deskewed}")
|
||||
|
||||
if dir_in:
|
||||
self.ls_imgs = os.listdir(dir_in)
|
||||
self.ls_imgs = [ind_img for ind_img in self.ls_imgs if ind_img.endswith('.jpg') or ind_img.endswith('.jpeg') or ind_img.endswith('.png') or ind_img.endswith('.tif') or ind_img.endswith('.tiff') or ind_img.endswith('.JPG') or ind_img.endswith('.JPEG') or ind_img.endswith('.TIF') or ind_img.endswith('.TIFF') or ind_img.endswith('.PNG')]
|
||||
ls_imgs = [os.path.join(dir_in, image_filename)
|
||||
for image_filename in filter(is_image_filename,
|
||||
os.listdir(dir_in))]
|
||||
elif image_filename:
|
||||
self.ls_imgs = [image_filename]
|
||||
ls_imgs = [image_filename]
|
||||
else:
|
||||
raise ValueError("run requires either a single image filename or a directory")
|
||||
|
||||
for img_filename in self.ls_imgs:
|
||||
for img_filename in ls_imgs:
|
||||
self.logger.info(img_filename)
|
||||
t0 = time.time()
|
||||
|
||||
self.reset_file_name_dir(os.path.join(dir_in or "", img_filename))
|
||||
self.reset_file_name_dir(img_filename, dir_out)
|
||||
if self.enable_plotting:
|
||||
self.plotter = EynollahPlotter(dir_out=dir_out,
|
||||
dir_of_all=dir_of_all,
|
||||
dir_save_page=dir_save_page,
|
||||
dir_of_deskewed=dir_of_deskewed,
|
||||
dir_of_cropped_images=dir_of_cropped_images,
|
||||
dir_of_layout=dir_of_layout,
|
||||
image_filename_stem=Path(image_filename).stem)
|
||||
#print("text region early -11 in %.1fs", time.time() - t0)
|
||||
if os.path.exists(self.writer.output_filename):
|
||||
if overwrite:
|
||||
|
|
@ -5236,19 +5242,6 @@ class Eynollah:
|
|||
|
||||
self.logger.info("Step 5/5: Output Generation")
|
||||
|
||||
output_config = []
|
||||
if self.enable_plotting:
|
||||
output_config.append("Saving debug plots")
|
||||
if self.dir_of_cropped_images:
|
||||
output_config.append(f"Saving cropped images to: {self.dir_of_cropped_images}")
|
||||
if self.dir_of_layout:
|
||||
output_config.append(f"Saving layout plots to: {self.dir_of_layout}")
|
||||
if self.dir_of_deskewed:
|
||||
output_config.append(f"Saving deskewed images to: {self.dir_of_deskewed}")
|
||||
|
||||
if output_config:
|
||||
self.logger.info("Output configuration:\n * %s", "\n * ".join(output_config))
|
||||
|
||||
pcgts = self.writer.build_pagexml_full_layout(
|
||||
contours_only_text_parent, contours_only_text_parent_h, page_coord, order_text_new, id_of_texts_tot,
|
||||
all_found_textline_polygons, all_found_textline_polygons_h, all_box_coord, all_box_coord_h,
|
||||
|
|
@ -5368,21 +5361,8 @@ class Eynollah:
|
|||
self.logger.info(f"Detection of reading order took {time.time() - t_order:.1f}s")
|
||||
|
||||
self.logger.info("Step 5/5: Output Generation")
|
||||
|
||||
self.logger.info("Generating PAGE-XML output")
|
||||
|
||||
if self.enable_plotting:
|
||||
self.logger.info("Saving debug plots")
|
||||
|
||||
if self.dir_of_cropped_images:
|
||||
self.logger.info(f"Saving cropped images to: {self.dir_of_cropped_images}")
|
||||
|
||||
if self.dir_of_layout:
|
||||
self.logger.info(f"Saving layout plots to: {self.dir_of_layout}")
|
||||
|
||||
if self.dir_of_deskewed:
|
||||
self.logger.info(f"Saving deskewed images to: {self.dir_of_deskewed}")
|
||||
|
||||
pcgts = self.writer.build_pagexml_no_full_layout(
|
||||
txt_con_org, page_coord, order_text_new, id_of_texts_tot,
|
||||
all_found_textline_polygons, all_box_coord, polygons_of_images, polygons_of_marginals_left, polygons_of_marginals_right,
|
||||
|
|
@ -5400,32 +5380,19 @@ class Eynollah_ocr:
|
|||
dir_models,
|
||||
model_name=None,
|
||||
dir_xmls=None,
|
||||
dir_in=None,
|
||||
image_filename=None,
|
||||
dir_in_bin=None,
|
||||
dir_out=None,
|
||||
dir_out_image_text=None,
|
||||
tr_ocr=False,
|
||||
batch_size=None,
|
||||
export_textline_images_and_text=False,
|
||||
do_not_mask_with_textline_contour=False,
|
||||
prediction_with_both_of_rgb_and_bin=False,
|
||||
pref_of_dataset=None,
|
||||
min_conf_value_of_textline_text : Optional[float]=None,
|
||||
logger=None,
|
||||
):
|
||||
self.dir_in = dir_in
|
||||
self.image_filename = image_filename
|
||||
self.dir_in_bin = dir_in_bin
|
||||
self.dir_out = dir_out
|
||||
self.dir_xmls = dir_xmls
|
||||
self.dir_models = dir_models
|
||||
self.model_name = model_name
|
||||
self.tr_ocr = tr_ocr
|
||||
self.export_textline_images_and_text = export_textline_images_and_text
|
||||
self.do_not_mask_with_textline_contour = do_not_mask_with_textline_contour
|
||||
self.dir_out_image_text = dir_out_image_text
|
||||
self.prediction_with_both_of_rgb_and_bin = prediction_with_both_of_rgb_and_bin
|
||||
self.pref_of_dataset = pref_of_dataset
|
||||
self.logger = logger if logger else getLogger('eynollah')
|
||||
|
||||
|
|
@ -5477,24 +5444,27 @@ class Eynollah_ocr:
|
|||
)
|
||||
self.end_character = len(characters) + 2
|
||||
|
||||
def run(self, overwrite : bool = False):
|
||||
if self.dir_in:
|
||||
ls_imgs = os.listdir(self.dir_in)
|
||||
ls_imgs = [ind_img for ind_img in ls_imgs if ind_img.endswith('.jpg') or ind_img.endswith('.jpeg') or ind_img.endswith('.png') or ind_img.endswith('.tif') or ind_img.endswith('.tiff') or ind_img.endswith('.JPG') or ind_img.endswith('.JPEG') or ind_img.endswith('.TIF') or ind_img.endswith('.TIFF') or ind_img.endswith('.PNG')]
|
||||
def run(self, overwrite: bool = False,
|
||||
dir_in: Optional[str] = None,
|
||||
dir_in_bin: Optional[str] = None,
|
||||
image_filename: Optional[str] = None,
|
||||
dir_xmls: Optional[str] = None,
|
||||
dir_out_image_text: Optional[str] = None,
|
||||
dir_out: Optional[str] = None,
|
||||
):
|
||||
if dir_in:
|
||||
ls_imgs = [os.path.join(dir_in, image_filename)
|
||||
for image_filename in filter(is_image_filename,
|
||||
os.listdir(dir_in))]
|
||||
else:
|
||||
ls_imgs = [self.image_filename]
|
||||
ls_imgs = [image_filename]
|
||||
|
||||
if self.tr_ocr:
|
||||
tr_ocr_input_height_and_width = 384
|
||||
for ind_img in ls_imgs:
|
||||
if self.dir_in:
|
||||
file_name = Path(ind_img).stem
|
||||
dir_img = os.path.join(self.dir_in, ind_img)
|
||||
else:
|
||||
file_name = Path(self.image_filename).stem
|
||||
dir_img = self.image_filename
|
||||
dir_xml = os.path.join(self.dir_xmls, file_name+'.xml')
|
||||
out_file_ocr = os.path.join(self.dir_out, file_name+'.xml')
|
||||
for dir_img in ls_imgs:
|
||||
file_name = Path(dir_img).stem
|
||||
dir_xml = os.path.join(dir_xmls, file_name+'.xml')
|
||||
out_file_ocr = os.path.join(dir_out, file_name+'.xml')
|
||||
|
||||
if os.path.exists(out_file_ocr):
|
||||
if overwrite:
|
||||
|
|
@ -5505,8 +5475,8 @@ class Eynollah_ocr:
|
|||
|
||||
img = cv2.imread(dir_img)
|
||||
|
||||
if self.dir_out_image_text:
|
||||
out_image_with_text = os.path.join(self.dir_out_image_text, file_name+'.png')
|
||||
if dir_out_image_text:
|
||||
out_image_with_text = os.path.join(dir_out_image_text, file_name+'.png')
|
||||
image_text = Image.new("RGB", (img.shape[1], img.shape[0]), "white")
|
||||
draw = ImageDraw.Draw(image_text)
|
||||
total_bb_coordinates = []
|
||||
|
|
@ -5544,7 +5514,7 @@ class Eynollah_ocr:
|
|||
textline_coords = np.array( [ [ int(x.split(',')[0]) , int(x.split(',')[1]) ] for x in p_h] )
|
||||
x,y,w,h = cv2.boundingRect(textline_coords)
|
||||
|
||||
if self.dir_out_image_text:
|
||||
if dir_out_image_text:
|
||||
total_bb_coordinates.append([x,y,w,h])
|
||||
|
||||
h2w_ratio = h/float(w)
|
||||
|
|
@ -5557,7 +5527,7 @@ class Eynollah_ocr:
|
|||
img_crop = img_poly_on_img[y:y+h, x:x+w, :]
|
||||
img_crop[mask_poly==0] = 255
|
||||
|
||||
|
||||
self.logger.debug("processing %d lines for '%s'", len(cropped_lines), nn.attrib['id'])
|
||||
if h2w_ratio > 0.1:
|
||||
cropped_lines.append(resize_image(img_crop, tr_ocr_input_height_and_width, tr_ocr_input_height_and_width) )
|
||||
cropped_lines_meging_indexing.append(0)
|
||||
|
|
@ -5666,7 +5636,7 @@ class Eynollah_ocr:
|
|||
|
||||
unique_cropped_lines_region_indexer = np.unique(cropped_lines_region_indexer)
|
||||
|
||||
if self.dir_out_image_text:
|
||||
if dir_out_image_text:
|
||||
|
||||
font_path = "Charis-7.000/Charis-Regular.ttf" # Make sure this file exists!
|
||||
font = ImageFont.truetype(font_path, 40)
|
||||
|
|
@ -5794,18 +5764,10 @@ class Eynollah_ocr:
|
|||
|
||||
img_size=(image_width, image_height)
|
||||
|
||||
for ind_img in ls_imgs:
|
||||
if self.dir_in:
|
||||
file_name = Path(ind_img).stem
|
||||
dir_img = os.path.join(self.dir_in, ind_img)
|
||||
else:
|
||||
file_name = Path(self.image_filename).stem
|
||||
dir_img = self.image_filename
|
||||
|
||||
#file_name = Path(ind_img).stem
|
||||
#dir_img = os.path.join(self.dir_in, ind_img)
|
||||
dir_xml = os.path.join(self.dir_xmls, file_name+'.xml')
|
||||
out_file_ocr = os.path.join(self.dir_out, file_name+'.xml')
|
||||
for dir_img in ls_imgs:
|
||||
file_name = Path(dir_img).stem
|
||||
dir_xml = os.path.join(dir_xmls, file_name+'.xml')
|
||||
out_file_ocr = os.path.join(dir_out, file_name+'.xml')
|
||||
|
||||
if os.path.exists(out_file_ocr):
|
||||
if overwrite:
|
||||
|
|
@ -5815,13 +5777,13 @@ class Eynollah_ocr:
|
|||
continue
|
||||
|
||||
img = cv2.imread(dir_img)
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
cropped_lines_bin = []
|
||||
dir_img_bin = os.path.join(self.dir_in_bin, file_name+'.png')
|
||||
dir_img_bin = os.path.join(dir_in_bin, file_name+'.png')
|
||||
img_bin = cv2.imread(dir_img_bin)
|
||||
|
||||
if self.dir_out_image_text:
|
||||
out_image_with_text = os.path.join(self.dir_out_image_text, file_name+'.png')
|
||||
if dir_out_image_text:
|
||||
out_image_with_text = os.path.join(dir_out_image_text, file_name+'.png')
|
||||
image_text = Image.new("RGB", (img.shape[1], img.shape[0]), "white")
|
||||
draw = ImageDraw.Draw(image_text)
|
||||
total_bb_coordinates = []
|
||||
|
|
@ -5865,13 +5827,13 @@ class Eynollah_ocr:
|
|||
if type_textregion=='drop-capital':
|
||||
angle_degrees = 0
|
||||
|
||||
if self.dir_out_image_text:
|
||||
if dir_out_image_text:
|
||||
total_bb_coordinates.append([x,y,w,h])
|
||||
|
||||
w_scaled = w * image_height/float(h)
|
||||
|
||||
img_poly_on_img = np.copy(img)
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
img_poly_on_img_bin = np.copy(img_bin)
|
||||
img_crop_bin = img_poly_on_img_bin[y:y+h, x:x+w, :]
|
||||
|
||||
|
|
@ -5894,7 +5856,7 @@ class Eynollah_ocr:
|
|||
|
||||
img_crop = rotate_image_with_padding(img_crop, better_des_slope )
|
||||
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
img_crop_bin = rotate_image_with_padding(img_crop_bin, better_des_slope )
|
||||
|
||||
mask_poly = rotate_image_with_padding(mask_poly, better_des_slope )
|
||||
|
|
@ -5909,13 +5871,13 @@ class Eynollah_ocr:
|
|||
if not self.do_not_mask_with_textline_contour:
|
||||
img_crop[mask_poly==0] = 255
|
||||
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
img_crop_bin = img_crop_bin[y_n:y_n+h_n, x_n:x_n+w_n, :]
|
||||
if not self.do_not_mask_with_textline_contour:
|
||||
img_crop_bin[mask_poly==0] = 255
|
||||
|
||||
if mask_poly[:,:,0].sum() /float(w_n*h_n) < 0.50 and w_scaled > 90:
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
img_crop, img_crop_bin = break_curved_line_into_small_pieces_and_then_merge(img_crop, mask_poly, img_crop_bin)
|
||||
else:
|
||||
img_crop, _ = break_curved_line_into_small_pieces_and_then_merge(img_crop, mask_poly)
|
||||
|
|
@ -5925,14 +5887,14 @@ class Eynollah_ocr:
|
|||
better_des_slope = 0
|
||||
if not self.do_not_mask_with_textline_contour:
|
||||
img_crop[mask_poly==0] = 255
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
if not self.do_not_mask_with_textline_contour:
|
||||
img_crop_bin[mask_poly==0] = 255
|
||||
if type_textregion=='drop-capital':
|
||||
pass
|
||||
else:
|
||||
if mask_poly[:,:,0].sum() /float(w*h) < 0.50 and w_scaled > 90:
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
img_crop, img_crop_bin = break_curved_line_into_small_pieces_and_then_merge(img_crop, mask_poly, img_crop_bin)
|
||||
else:
|
||||
img_crop, _ = break_curved_line_into_small_pieces_and_then_merge(img_crop, mask_poly)
|
||||
|
|
@ -5947,14 +5909,12 @@ class Eynollah_ocr:
|
|||
cropped_lines_ver_index.append(0)
|
||||
|
||||
cropped_lines_meging_indexing.append(0)
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
img_fin = preprocess_and_resize_image_for_ocrcnn_model(img_crop_bin, image_height, image_width)
|
||||
cropped_lines_bin.append(img_fin)
|
||||
else:
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
splited_images, splited_images_bin = return_textlines_split_if_needed(img_crop, img_crop_bin, prediction_with_both_of_rgb_and_bin=self.prediction_with_both_of_rgb_and_bin)
|
||||
else:
|
||||
splited_images, splited_images_bin = return_textlines_split_if_needed(img_crop, None)
|
||||
splited_images, splited_images_bin = return_textlines_split_if_needed(
|
||||
img_crop, img_crop_bin if dir_in_bin is not None else None)
|
||||
if splited_images:
|
||||
img_fin = preprocess_and_resize_image_for_ocrcnn_model(splited_images[0], image_height, image_width)
|
||||
cropped_lines.append(img_fin)
|
||||
|
|
@ -5975,7 +5935,7 @@ class Eynollah_ocr:
|
|||
else:
|
||||
cropped_lines_ver_index.append(0)
|
||||
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
img_fin = preprocess_and_resize_image_for_ocrcnn_model(splited_images_bin[0], image_height, image_width)
|
||||
cropped_lines_bin.append(img_fin)
|
||||
img_fin = preprocess_and_resize_image_for_ocrcnn_model(splited_images_bin[1], image_height, image_width)
|
||||
|
|
@ -5991,7 +5951,7 @@ class Eynollah_ocr:
|
|||
else:
|
||||
cropped_lines_ver_index.append(0)
|
||||
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
img_fin = preprocess_and_resize_image_for_ocrcnn_model(img_crop_bin, image_height, image_width)
|
||||
cropped_lines_bin.append(img_fin)
|
||||
|
||||
|
|
@ -6004,29 +5964,15 @@ class Eynollah_ocr:
|
|||
if cheild_text.tag.endswith("Unicode"):
|
||||
textline_text = cheild_text.text
|
||||
if textline_text:
|
||||
if self.do_not_mask_with_textline_contour:
|
||||
if self.pref_of_dataset:
|
||||
with open(os.path.join(self.dir_out, file_name+'_line_'+str(indexer_textlines)+'_'+self.pref_of_dataset+'.txt'), 'w') as text_file:
|
||||
text_file.write(textline_text)
|
||||
|
||||
cv2.imwrite(os.path.join(self.dir_out, file_name+'_line_'+str(indexer_textlines)+'_'+self.pref_of_dataset+'.png'), img_crop )
|
||||
else:
|
||||
with open(os.path.join(self.dir_out, file_name+'_line_'+str(indexer_textlines)+'.txt'), 'w') as text_file:
|
||||
text_file.write(textline_text)
|
||||
|
||||
cv2.imwrite(os.path.join(self.dir_out, file_name+'_line_'+str(indexer_textlines)+'.png'), img_crop )
|
||||
else:
|
||||
if self.pref_of_dataset:
|
||||
with open(os.path.join(self.dir_out, file_name+'_line_'+str(indexer_textlines)+'_'+self.pref_of_dataset+'_masked.txt'), 'w') as text_file:
|
||||
text_file.write(textline_text)
|
||||
|
||||
cv2.imwrite(os.path.join(self.dir_out, file_name+'_line_'+str(indexer_textlines)+'_'+self.pref_of_dataset+'_masked.png'), img_crop )
|
||||
else:
|
||||
with open(os.path.join(self.dir_out, file_name+'_line_'+str(indexer_textlines)+'_masked.txt'), 'w') as text_file:
|
||||
text_file.write(textline_text)
|
||||
|
||||
cv2.imwrite(os.path.join(self.dir_out, file_name+'_line_'+str(indexer_textlines)+'_masked.png'), img_crop )
|
||||
base_name = os.path.join(dir_out, file_name + '_line_' + str(indexer_textlines))
|
||||
if self.pref_of_dataset:
|
||||
base_name += '_' + self.pref_of_dataset
|
||||
if not self.do_not_mask_with_textline_contour:
|
||||
base_name += '_masked'
|
||||
|
||||
with open(base_name + '.txt', 'w') as text_file:
|
||||
text_file.write(textline_text)
|
||||
cv2.imwrite(base_name + '.png', img_crop)
|
||||
indexer_textlines+=1
|
||||
|
||||
if not self.export_textline_images_and_text:
|
||||
|
|
@ -6057,7 +6003,7 @@ class Eynollah_ocr:
|
|||
else:
|
||||
imgs_ver_flipped = None
|
||||
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
imgs_bin = cropped_lines_bin[n_start:]
|
||||
imgs_bin = np.array(imgs_bin)
|
||||
imgs_bin = imgs_bin.reshape(imgs_bin.shape[0], image_height, image_width, 3)
|
||||
|
|
@ -6087,7 +6033,7 @@ class Eynollah_ocr:
|
|||
imgs_ver_flipped = None
|
||||
|
||||
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
imgs_bin = cropped_lines_bin[n_start:n_end]
|
||||
imgs_bin = np.array(imgs_bin).reshape(self.b_s, image_height, image_width, 3)
|
||||
|
||||
|
|
@ -6100,6 +6046,7 @@ class Eynollah_ocr:
|
|||
imgs_bin_ver_flipped = None
|
||||
|
||||
|
||||
self.logger.debug("processing next %d lines", len(imgs))
|
||||
preds = self.prediction_model.predict(imgs, verbose=0)
|
||||
|
||||
if len(indices_ver)>0:
|
||||
|
|
@ -6126,7 +6073,7 @@ class Eynollah_ocr:
|
|||
if len(indices_where_flipped_conf_value_is_higher)>0:
|
||||
indices_to_be_replaced = indices_ver[indices_where_flipped_conf_value_is_higher]
|
||||
preds[indices_to_be_replaced,:,:] = preds_flipped[indices_where_flipped_conf_value_is_higher, :, :]
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
preds_bin = self.prediction_model.predict(imgs_bin, verbose=0)
|
||||
|
||||
if len(indices_ver)>0:
|
||||
|
|
@ -6173,7 +6120,7 @@ class Eynollah_ocr:
|
|||
extracted_texts.append("")
|
||||
extracted_conf_value.append(0)
|
||||
del cropped_lines
|
||||
if self.prediction_with_both_of_rgb_and_bin:
|
||||
if dir_in_bin is not None:
|
||||
del cropped_lines_bin
|
||||
gc.collect()
|
||||
|
||||
|
|
@ -6186,7 +6133,7 @@ class Eynollah_ocr:
|
|||
unique_cropped_lines_region_indexer = np.unique(cropped_lines_region_indexer)
|
||||
|
||||
|
||||
if self.dir_out_image_text:
|
||||
if dir_out_image_text:
|
||||
|
||||
font_path = "Charis-7.000/Charis-Regular.ttf" # Make sure this file exists!
|
||||
font = ImageFont.truetype(font_path, 40)
|
||||
|
|
|
|||
|
|
@ -21,6 +21,7 @@ from tensorflow.keras.models import load_model
|
|||
from .utils.resize import resize_image
|
||||
from .utils.pil_cv2 import pil2cv
|
||||
from .utils import (
|
||||
is_image_filename,
|
||||
crop_image_inside_box
|
||||
)
|
||||
|
||||
|
|
@ -701,13 +702,13 @@ class Enhancer:
|
|||
t0_tot = time.time()
|
||||
|
||||
if dir_in:
|
||||
self.ls_imgs = os.listdir(dir_in)
|
||||
ls_imgs = list(filter(is_image_filename, os.listdir(dir_in)))
|
||||
elif image_filename:
|
||||
self.ls_imgs = [image_filename]
|
||||
ls_imgs = [image_filename]
|
||||
else:
|
||||
raise ValueError("run requires either a single image filename or a directory")
|
||||
|
||||
for img_filename in self.ls_imgs:
|
||||
for img_filename in ls_imgs:
|
||||
self.logger.info(img_filename)
|
||||
t0 = time.time()
|
||||
|
||||
|
|
|
|||
|
|
@ -25,6 +25,7 @@ from .utils.contour import (
|
|||
return_contours_of_image,
|
||||
return_parent_contours,
|
||||
)
|
||||
from .utils import is_xml_filename
|
||||
|
||||
DPI_THRESHOLD = 298
|
||||
KERNEL = np.ones((5, 5), np.uint8)
|
||||
|
|
@ -39,7 +40,7 @@ class machine_based_reading_order_on_layout:
|
|||
):
|
||||
self.dir_out = dir_out
|
||||
|
||||
self.logger = logger if logger else getLogger('mbro on layout')
|
||||
self.logger = logger if logger else getLogger('mbreorder')
|
||||
# for parallelization of CPU-intensive tasks:
|
||||
self.executor = ProcessPoolExecutor(max_workers=cpu_count(), timeout=1200)
|
||||
atexit.register(self.executor.shutdown)
|
||||
|
|
@ -751,13 +752,13 @@ class machine_based_reading_order_on_layout:
|
|||
t0_tot = time.time()
|
||||
|
||||
if dir_in:
|
||||
self.ls_xmls = os.listdir(dir_in)
|
||||
ls_xmls = list(filter(is_xml_filename, os.listdir(dir_in)))
|
||||
elif xml_filename:
|
||||
self.ls_xmls = [xml_filename]
|
||||
ls_xmls = [xml_filename]
|
||||
else:
|
||||
raise ValueError("run requires either a single image filename or a directory")
|
||||
|
||||
for xml_filename in self.ls_xmls:
|
||||
for xml_filename in ls_xmls:
|
||||
self.logger.info(xml_filename)
|
||||
t0 = time.time()
|
||||
|
||||
|
|
|
|||
|
|
@ -16,6 +16,7 @@ import tensorflow as tf
|
|||
from tensorflow.keras.models import load_model
|
||||
from tensorflow.python.keras import backend as tensorflow_backend
|
||||
|
||||
from .utils import is_image_filename
|
||||
|
||||
def resize_image(img_in, input_height, input_width):
|
||||
return cv2.resize(img_in, (input_width, input_height), interpolation=cv2.INTER_NEAREST)
|
||||
|
|
@ -347,7 +348,7 @@ class SbbBinarizer:
|
|||
cv2.imwrite(output, img_last)
|
||||
return img_last
|
||||
else:
|
||||
ls_imgs = os.listdir(dir_in)
|
||||
ls_imgs = list(filter(is_image_filename, os.listdir(dir_in)))
|
||||
for image_name in ls_imgs:
|
||||
image_stem = image_name.split('.')[0]
|
||||
print(image_name,'image_name')
|
||||
|
|
|
|||
|
|
@ -2194,3 +2194,14 @@ def return_boxes_of_images_by_order_of_reading_new(
|
|||
return boxes, peaks_neg_tot_tables_new
|
||||
else:
|
||||
return boxes, peaks_neg_tot_tables
|
||||
|
||||
def is_image_filename(fname: str) -> bool:
|
||||
return fname.lower().endswith(('.jpg',
|
||||
'.jpeg',
|
||||
'.png',
|
||||
'.tif',
|
||||
'.tiff',
|
||||
))
|
||||
|
||||
def is_xml_filename(fname: str) -> bool:
|
||||
return fname.lower().endswith('.xml')
|
||||
|
|
|
|||
|
|
@ -109,13 +109,13 @@ def fit_text_single_line(draw, text, font_path, max_width, max_height):
|
|||
|
||||
return ImageFont.truetype(font_path, 10) # Smallest font fallback
|
||||
|
||||
def return_textlines_split_if_needed(textline_image, textline_image_bin, prediction_with_both_of_rgb_and_bin=False):
|
||||
def return_textlines_split_if_needed(textline_image, textline_image_bin=None):
|
||||
|
||||
split_point = return_start_and_end_of_common_text_of_textline_ocr_without_common_section(textline_image)
|
||||
if split_point:
|
||||
image1 = textline_image[:, :split_point,:]# image.crop((0, 0, width2, height))
|
||||
image2 = textline_image[:, split_point:,:]#image.crop((width1, 0, width, height))
|
||||
if prediction_with_both_of_rgb_and_bin:
|
||||
if textline_image_bin is not None:
|
||||
image1_bin = textline_image_bin[:, :split_point,:]# image.crop((0, 0, width2, height))
|
||||
image2_bin = textline_image_bin[:, split_point:,:]#image.crop((width1, 0, width, height))
|
||||
return [image1, image2], [image1_bin, image2_bin]
|
||||
|
|
|
|||
1626
tests/resources/euler_rechenkunst01_1738_0025.xml
Normal file
1626
tests/resources/euler_rechenkunst01_1738_0025.xml
Normal file
File diff suppressed because it is too large
Load diff
2129
tests/resources/kant_aufklaerung_1784_0020.xml
Normal file
2129
tests/resources/kant_aufklaerung_1784_0020.xml
Normal file
File diff suppressed because it is too large
Load diff
|
|
@ -2,7 +2,13 @@ from os import environ
|
|||
from pathlib import Path
|
||||
import logging
|
||||
from PIL import Image
|
||||
from eynollah.cli import layout as layout_cli, binarization as binarization_cli
|
||||
from eynollah.cli import (
|
||||
layout as layout_cli,
|
||||
binarization as binarization_cli,
|
||||
enhancement as enhancement_cli,
|
||||
machine_based_reading_order as mbreorder_cli,
|
||||
ocr as ocr_cli,
|
||||
)
|
||||
from click.testing import CliRunner
|
||||
from ocrd_modelfactory import page_from_file
|
||||
from ocrd_models.constants import NAMESPACES as NS
|
||||
|
|
@ -44,8 +50,7 @@ def test_run_eynollah_layout_filename(tmp_path, subtests, pytestconfig, caplog):
|
|||
options=options):
|
||||
with caplog.filtering(only_eynollah):
|
||||
result = runner.invoke(layout_cli, args + options, catch_exceptions=False)
|
||||
print(result)
|
||||
assert result.exit_code == 0
|
||||
assert result.exit_code == 0, result.stdout
|
||||
logmsgs = [logrec.message for logrec in caplog.records]
|
||||
assert str(infile) in logmsgs
|
||||
assert outfile.exists()
|
||||
|
|
@ -72,9 +77,8 @@ def test_run_eynollah_layout_directory(tmp_path, pytestconfig, caplog):
|
|||
return logrec.name == 'eynollah'
|
||||
runner = CliRunner()
|
||||
with caplog.filtering(only_eynollah):
|
||||
result = runner.invoke(layout_cli, args)
|
||||
print(result)
|
||||
assert result.exit_code == 0
|
||||
result = runner.invoke(layout_cli, args, catch_exceptions=False)
|
||||
assert result.exit_code == 0, result.stdout
|
||||
logmsgs = [logrec.message for logrec in caplog.records]
|
||||
assert len([logmsg for logmsg in logmsgs if logmsg.startswith('Job done in')]) == 2
|
||||
assert any(logmsg for logmsg in logmsgs if logmsg.startswith('All jobs done in'))
|
||||
|
|
@ -88,6 +92,8 @@ def test_run_eynollah_binarization_filename(tmp_path, subtests, pytestconfig, ca
|
|||
'-i', str(infile),
|
||||
'-o', str(outfile),
|
||||
]
|
||||
if pytestconfig.getoption('verbose') > 0:
|
||||
args.extend(['-l', 'DEBUG'])
|
||||
caplog.set_level(logging.INFO)
|
||||
def only_eynollah(logrec):
|
||||
return logrec.name == 'SbbBinarizer'
|
||||
|
|
@ -99,9 +105,8 @@ def test_run_eynollah_binarization_filename(tmp_path, subtests, pytestconfig, ca
|
|||
with subtests.test(#msg="test CLI",
|
||||
options=options):
|
||||
with caplog.filtering(only_eynollah):
|
||||
result = runner.invoke(binarization_cli, args + options)
|
||||
print(result)
|
||||
assert result.exit_code == 0
|
||||
result = runner.invoke(binarization_cli, args + options, catch_exceptions=False)
|
||||
assert result.exit_code == 0, result.stdout
|
||||
logmsgs = [logrec.message for logrec in caplog.records]
|
||||
assert any(True for logmsg in logmsgs if logmsg.startswith('Predicting'))
|
||||
assert outfile.exists()
|
||||
|
|
@ -119,14 +124,186 @@ def test_run_eynollah_binarization_directory(tmp_path, subtests, pytestconfig, c
|
|||
'-di', str(indir),
|
||||
'-o', str(outdir),
|
||||
]
|
||||
if pytestconfig.getoption('verbose') > 0:
|
||||
args.extend(['-l', 'DEBUG'])
|
||||
caplog.set_level(logging.INFO)
|
||||
def only_eynollah(logrec):
|
||||
return logrec.name == 'SbbBinarizer'
|
||||
runner = CliRunner()
|
||||
with caplog.filtering(only_eynollah):
|
||||
result = runner.invoke(binarization_cli, args)
|
||||
print(result)
|
||||
assert result.exit_code == 0
|
||||
result = runner.invoke(binarization_cli, args, catch_exceptions=False)
|
||||
assert result.exit_code == 0, result.stdout
|
||||
logmsgs = [logrec.message for logrec in caplog.records]
|
||||
assert len([logmsg for logmsg in logmsgs if logmsg.startswith('Predicting')]) == 2
|
||||
assert len(list(outdir.iterdir())) == 2
|
||||
|
||||
def test_run_eynollah_enhancement_filename(tmp_path, subtests, pytestconfig, caplog):
|
||||
infile = testdir.joinpath('resources/kant_aufklaerung_1784_0020.tif')
|
||||
outfile = tmp_path.joinpath('kant_aufklaerung_1784_0020.png')
|
||||
args = [
|
||||
'-m', EYNOLLAH_MODELS,
|
||||
'-i', str(infile),
|
||||
'-o', str(outfile.parent),
|
||||
# subtests write to same location
|
||||
'--overwrite',
|
||||
]
|
||||
if pytestconfig.getoption('verbose') > 0:
|
||||
args.extend(['-l', 'DEBUG'])
|
||||
caplog.set_level(logging.INFO)
|
||||
def only_eynollah(logrec):
|
||||
return logrec.name == 'enhancement'
|
||||
runner = CliRunner()
|
||||
for options in [
|
||||
[], # defaults
|
||||
["-sos"],
|
||||
]:
|
||||
with subtests.test(#msg="test CLI",
|
||||
options=options):
|
||||
with caplog.filtering(only_eynollah):
|
||||
result = runner.invoke(enhancement_cli, args + options, catch_exceptions=False)
|
||||
assert result.exit_code == 0, result.stdout
|
||||
logmsgs = [logrec.message for logrec in caplog.records]
|
||||
assert any(True for logmsg in logmsgs if logmsg.startswith('Image was enhanced')), logmsgs
|
||||
assert outfile.exists()
|
||||
with Image.open(infile) as original_img:
|
||||
original_size = original_img.size
|
||||
with Image.open(outfile) as enhanced_img:
|
||||
enhanced_size = enhanced_img.size
|
||||
assert (original_size == enhanced_size) == ("-sos" in options)
|
||||
|
||||
def test_run_eynollah_enhancement_directory(tmp_path, subtests, pytestconfig, caplog):
|
||||
indir = testdir.joinpath('resources')
|
||||
outdir = tmp_path
|
||||
args = [
|
||||
'-m', EYNOLLAH_MODELS,
|
||||
'-di', str(indir),
|
||||
'-o', str(outdir),
|
||||
]
|
||||
if pytestconfig.getoption('verbose') > 0:
|
||||
args.extend(['-l', 'DEBUG'])
|
||||
caplog.set_level(logging.INFO)
|
||||
def only_eynollah(logrec):
|
||||
return logrec.name == 'enhancement'
|
||||
runner = CliRunner()
|
||||
with caplog.filtering(only_eynollah):
|
||||
result = runner.invoke(enhancement_cli, args, catch_exceptions=False)
|
||||
assert result.exit_code == 0, result.stdout
|
||||
logmsgs = [logrec.message for logrec in caplog.records]
|
||||
assert len([logmsg for logmsg in logmsgs if logmsg.startswith('Image was enhanced')]) == 2
|
||||
assert len(list(outdir.iterdir())) == 2
|
||||
|
||||
def test_run_eynollah_mbreorder_filename(tmp_path, subtests, pytestconfig, caplog):
|
||||
infile = testdir.joinpath('resources/kant_aufklaerung_1784_0020.xml')
|
||||
outfile = tmp_path.joinpath('kant_aufklaerung_1784_0020.xml')
|
||||
args = [
|
||||
'-m', EYNOLLAH_MODELS,
|
||||
'-i', str(infile),
|
||||
'-o', str(outfile.parent),
|
||||
]
|
||||
if pytestconfig.getoption('verbose') > 0:
|
||||
args.extend(['-l', 'DEBUG'])
|
||||
caplog.set_level(logging.INFO)
|
||||
def only_eynollah(logrec):
|
||||
return logrec.name == 'mbreorder'
|
||||
runner = CliRunner()
|
||||
with caplog.filtering(only_eynollah):
|
||||
result = runner.invoke(mbreorder_cli, args, catch_exceptions=False)
|
||||
assert result.exit_code == 0, result.stdout
|
||||
logmsgs = [logrec.message for logrec in caplog.records]
|
||||
# FIXME: mbreorder has no logging!
|
||||
#assert any(True for logmsg in logmsgs if logmsg.startswith('???')), logmsgs
|
||||
assert outfile.exists()
|
||||
#in_tree = page_from_file(str(infile)).etree
|
||||
#in_order = in_tree.xpath("//page:OrderedGroup//@regionRef", namespaces=NS)
|
||||
out_tree = page_from_file(str(outfile)).etree
|
||||
out_order = out_tree.xpath("//page:OrderedGroup//@regionRef", namespaces=NS)
|
||||
#assert len(out_order) >= 2, "result is inaccurate"
|
||||
#assert in_order != out_order
|
||||
assert out_order == ['r_1_1', 'r_2_1', 'r_2_2', 'r_2_3']
|
||||
|
||||
def test_run_eynollah_mbreorder_directory(tmp_path, subtests, pytestconfig, caplog):
|
||||
indir = testdir.joinpath('resources')
|
||||
outdir = tmp_path
|
||||
args = [
|
||||
'-m', EYNOLLAH_MODELS,
|
||||
'-di', str(indir),
|
||||
'-o', str(outdir),
|
||||
]
|
||||
if pytestconfig.getoption('verbose') > 0:
|
||||
args.extend(['-l', 'DEBUG'])
|
||||
caplog.set_level(logging.INFO)
|
||||
def only_eynollah(logrec):
|
||||
return logrec.name == 'mbreorder'
|
||||
runner = CliRunner()
|
||||
with caplog.filtering(only_eynollah):
|
||||
result = runner.invoke(mbreorder_cli, args, catch_exceptions=False)
|
||||
assert result.exit_code == 0, result.stdout
|
||||
logmsgs = [logrec.message for logrec in caplog.records]
|
||||
# FIXME: mbreorder has no logging!
|
||||
#assert len([logmsg for logmsg in logmsgs if logmsg.startswith('???')]) == 2
|
||||
assert len(list(outdir.iterdir())) == 2
|
||||
|
||||
def test_run_eynollah_ocr_filename(tmp_path, subtests, pytestconfig, caplog):
|
||||
infile = testdir.joinpath('resources/kant_aufklaerung_1784_0020.tif')
|
||||
outfile = tmp_path.joinpath('kant_aufklaerung_1784_0020.xml')
|
||||
outrenderfile = tmp_path.joinpath('render').joinpath('kant_aufklaerung_1784_0020.xml')
|
||||
outrenderfile.parent.mkdir()
|
||||
args = [
|
||||
'-m', EYNOLLAH_MODELS,
|
||||
'-i', str(infile),
|
||||
'-dx', str(infile.parent),
|
||||
'-o', str(outfile.parent),
|
||||
# subtests write to same location
|
||||
'--overwrite',
|
||||
]
|
||||
if pytestconfig.getoption('verbose') > 0:
|
||||
args.extend(['-l', 'DEBUG'])
|
||||
caplog.set_level(logging.DEBUG)
|
||||
def only_eynollah(logrec):
|
||||
return logrec.name == 'eynollah'
|
||||
runner = CliRunner()
|
||||
for options in [
|
||||
[], # defaults
|
||||
["-doit", str(outrenderfile.parent)],
|
||||
["-trocr"],
|
||||
]:
|
||||
with subtests.test(#msg="test CLI",
|
||||
options=options):
|
||||
with caplog.filtering(only_eynollah):
|
||||
result = runner.invoke(ocr_cli, args + options, catch_exceptions=False)
|
||||
assert result.exit_code == 0, result.stdout
|
||||
logmsgs = [logrec.message for logrec in caplog.records]
|
||||
# FIXME: ocr has no logging!
|
||||
#assert any(True for logmsg in logmsgs if logmsg.startswith('???')), logmsgs
|
||||
assert outfile.exists()
|
||||
if "-doit" in options:
|
||||
assert outrenderfile.exists()
|
||||
#in_tree = page_from_file(str(infile)).etree
|
||||
#in_order = in_tree.xpath("//page:OrderedGroup//@regionRef", namespaces=NS)
|
||||
out_tree = page_from_file(str(outfile)).etree
|
||||
out_texts = out_tree.xpath("//page:TextLine/page:TextEquiv[last()]/page:Unicode/text()", namespaces=NS)
|
||||
assert len(out_texts) >= 2, ("result is inaccurate", out_texts)
|
||||
assert sum(map(len, out_texts)) > 100, ("result is inaccurate", out_texts)
|
||||
|
||||
def test_run_eynollah_ocr_directory(tmp_path, subtests, pytestconfig, caplog):
|
||||
indir = testdir.joinpath('resources')
|
||||
outdir = tmp_path
|
||||
args = [
|
||||
'-m', EYNOLLAH_MODELS,
|
||||
'-di', str(indir),
|
||||
'-dx', str(indir),
|
||||
'-o', str(outdir),
|
||||
]
|
||||
if pytestconfig.getoption('verbose') > 0:
|
||||
args.extend(['-l', 'DEBUG'])
|
||||
caplog.set_level(logging.INFO)
|
||||
def only_eynollah(logrec):
|
||||
return logrec.name == 'eynollah'
|
||||
runner = CliRunner()
|
||||
with caplog.filtering(only_eynollah):
|
||||
result = runner.invoke(ocr_cli, args, catch_exceptions=False)
|
||||
assert result.exit_code == 0, result.stdout
|
||||
logmsgs = [logrec.message for logrec in caplog.records]
|
||||
# FIXME: ocr has no logging!
|
||||
#assert any(True for logmsg in logmsgs if logmsg.startswith('???')), logmsgs
|
||||
assert len(list(outdir.iterdir())) == 2
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue