qurator-spk/eynollah
1
0
Fork 0
mirror of https://github.com/qurator-spk/eynollah.git synced 2025-10-15 19:09:58 +02:00
Code Issues Projects Releases Wiki Activity

fix identifier scope in layout OCR options (w/o full_layout)

Browse source
This commit is contained in:
Robert Sachunsky 2025-10-08 16:52:22 +02:00
parent 096def1e9d
commit 8a2d682e12
1 changed files with 11 additions and 22 deletions
Download patch file Download diff file Expand all files Collapse all files

33
src/eynollah/eynollah.py
View file

@ -4726,7 +4726,6 @@ class Eynollah:
self.plotter.write_images_into_directory(polygons_of_images, image_page) self.plotter.write_images_into_directory(polygons_of_images, image_page)
t_order = time.time() t_order = time.time()
#if self.full_layout:
self.logger.info("Step 4/5: Reading Order Detection") self.logger.info("Step 4/5: Reading Order Detection")
if self.reading_order_machine_based: if self.reading_order_machine_based:
@ -4749,46 +4748,41 @@ class Eynollah:
boxes_d, textline_mask_tot_d) boxes_d, textline_mask_tot_d)
self.logger.info(f"Detection of reading order took {time.time() - t_order:.1f}s") self.logger.info(f"Detection of reading order took {time.time() - t_order:.1f}s")
ocr_all_textlines = None
ocr_all_textlines_marginals_left = None
ocr_all_textlines_marginals_right = None
ocr_all_textlines_h = None
ocr_all_textlines_drop = None
if self.ocr: if self.ocr:
self.logger.info("Step 4.5/5: OCR Processing") self.logger.info("Step 4.5/5: OCR Processing")
if not self.tr: if not self.tr:
gc.collect() gc.collect()
if len(all_found_textline_polygons)>0: if len(all_found_textline_polygons):
ocr_all_textlines = return_rnn_cnn_ocr_of_given_textlines( ocr_all_textlines = return_rnn_cnn_ocr_of_given_textlines(
image_page, all_found_textline_polygons, all_box_coord, image_page, all_found_textline_polygons, all_box_coord,
self.prediction_model, self.b_s_ocr, self.num_to_char, self.textline_light, self.curved_line) self.prediction_model, self.b_s_ocr, self.num_to_char, self.textline_light, self.curved_line)
else:
ocr_all_textlines = None
if all_found_textline_polygons_marginals_left and len(all_found_textline_polygons_marginals_left)>0: if len(all_found_textline_polygons_marginals_left):
ocr_all_textlines_marginals_left = return_rnn_cnn_ocr_of_given_textlines( ocr_all_textlines_marginals_left = return_rnn_cnn_ocr_of_given_textlines(
image_page, all_found_textline_polygons_marginals_left, all_box_coord_marginals_left, image_page, all_found_textline_polygons_marginals_left, all_box_coord_marginals_left,
self.prediction_model, self.b_s_ocr, self.num_to_char, self.textline_light, self.curved_line) self.prediction_model, self.b_s_ocr, self.num_to_char, self.textline_light, self.curved_line)
else:
ocr_all_textlines_marginals_left = None
if all_found_textline_polygons_marginals_right and len(all_found_textline_polygons_marginals_right)>0: if len(all_found_textline_polygons_marginals_right):
ocr_all_textlines_marginals_right = return_rnn_cnn_ocr_of_given_textlines( ocr_all_textlines_marginals_right = return_rnn_cnn_ocr_of_given_textlines(
image_page, all_found_textline_polygons_marginals_right, all_box_coord_marginals_right, image_page, all_found_textline_polygons_marginals_right, all_box_coord_marginals_right,
self.prediction_model, self.b_s_ocr, self.num_to_char, self.textline_light, self.curved_line) self.prediction_model, self.b_s_ocr, self.num_to_char, self.textline_light, self.curved_line)
else:
ocr_all_textlines_marginals_right = None
if all_found_textline_polygons_h and len(all_found_textline_polygons)>0: if self.full_layout and len(all_found_textline_polygons):
ocr_all_textlines_h = return_rnn_cnn_ocr_of_given_textlines( ocr_all_textlines_h = return_rnn_cnn_ocr_of_given_textlines(
image_page, all_found_textline_polygons_h, all_box_coord_h, image_page, all_found_textline_polygons_h, all_box_coord_h,
self.prediction_model, self.b_s_ocr, self.num_to_char, self.textline_light, self.curved_line) self.prediction_model, self.b_s_ocr, self.num_to_char, self.textline_light, self.curved_line)
else:
ocr_all_textlines_h = None
if polygons_of_drop_capitals and len(polygons_of_drop_capitals)>0: if self.full_layout and len(polygons_of_drop_capitals):
ocr_all_textlines_drop = return_rnn_cnn_ocr_of_given_textlines( ocr_all_textlines_drop = return_rnn_cnn_ocr_of_given_textlines(
image_page, polygons_of_drop_capitals, np.zeros((len(polygons_of_drop_capitals), 4)), image_page, polygons_of_drop_capitals, np.zeros((len(polygons_of_drop_capitals), 4)),
self.prediction_model, self.b_s_ocr, self.num_to_char, self.textline_light, self.curved_line) self.prediction_model, self.b_s_ocr, self.num_to_char, self.textline_light, self.curved_line)
else:
ocr_all_textlines_drop = None
else: else:
if self.light_version: if self.light_version:
@ -4805,6 +4799,7 @@ class Eynollah:
ind_tot = 0 ind_tot = 0
#cv2.imwrite('./img_out.png', image_page) #cv2.imwrite('./img_out.png', image_page)
ocr_all_textlines = [] ocr_all_textlines = []
# FIXME: what about lines in marginals / headings / drop-capitals here?
for indexing, ind_poly_first in enumerate(all_found_textline_polygons): for indexing, ind_poly_first in enumerate(all_found_textline_polygons):
ocr_textline_in_textregion = [] ocr_textline_in_textregion = []
for indexing2, ind_poly in enumerate(ind_poly_first): for indexing2, ind_poly in enumerate(ind_poly_first):
@ -4840,12 +4835,6 @@ class Eynollah:
ocr_textline_in_textregion.append(text_ocr) ocr_textline_in_textregion.append(text_ocr)
ind_tot = ind_tot +1 ind_tot = ind_tot +1
ocr_all_textlines.append(ocr_textline_in_textregion) ocr_all_textlines.append(ocr_textline_in_textregion)
else:
ocr_all_textlines = None
ocr_all_textlines_marginals_left = None
ocr_all_textlines_marginals_right = None
ocr_all_textlines_h = None
ocr_all_textlines_drop = None
self.logger.info("Step 5/5: Output Generation") self.logger.info("Step 5/5: Output Generation")