qurator-spk/eynollah
1
0
Fork 0
mirror of https://github.com/qurator-spk/eynollah.git synced 2025-06-10 04:39:54 +02:00
Code Issues Projects Releases Wiki Activity

🎨 unncesssary if True

Browse source
This commit is contained in:
Konstantin Baierer 2021-02-23 17:11:32 +01:00
parent 68d5c0d523
commit a65caa4d25
1 changed files with 71 additions and 84 deletions
Download patch file Download diff file Expand all files Collapse all files

25
sbb_newspapers_org_image/eynollah.py
View file

@ -171,17 +171,9 @@ class eynollah:
if img.shape[1] < img_width_model: if img.shape[1] < img_width_model:
img = cv2.resize(img, (img_height_model, img.shape[0]), interpolation=cv2.INTER_NEAREST) img = cv2.resize(img, (img_height_model, img.shape[0]), interpolation=cv2.INTER_NEAREST)
margin = True
if margin:
kernel = np.ones((5, 5), np.uint8)
margin = int(0 * img_width_model) margin = int(0 * img_width_model)
width_mid = img_width_model - 2 * margin width_mid = img_width_model - 2 * margin
height_mid = img_height_model - 2 * margin height_mid = img_height_model - 2 * margin
img = img / float(255.0) img = img / float(255.0)
img_h = img.shape[0] img_h = img.shape[0]
@ -203,7 +195,6 @@ class eynollah:
else: else:
index_x_d = i * width_mid index_x_d = i * width_mid
index_x_u = index_x_d + img_width_model index_x_u = index_x_d + img_width_model
if j == 0: if j == 0:
index_y_d = j * height_mid index_y_d = j * height_mid
index_y_u = index_y_d + img_height_model index_y_u = index_y_d + img_height_model
@ -254,9 +245,6 @@ class eynollah:
prediction_true = prediction_true.astype(int) prediction_true = prediction_true.astype(int)
del model_enhancement
del session_enhancemnet
return prediction_true return prediction_true
def calculate_width_height_by_columns(self, img, num_col, width_early, label_p_pred): def calculate_width_height_by_columns(self, img, num_col, width_early, label_p_pred):
@ -1252,7 +1240,6 @@ class eynollah:
id_indexer_l = 0 id_indexer_l = 0
if len(found_polygons_text_region) > 0: if len(found_polygons_text_region) > 0:
self.xml_reading_order(page, order_of_texts, id_of_texts, id_of_marginalia, found_polygons_marginals) self.xml_reading_order(page, order_of_texts, id_of_texts, id_of_marginalia, found_polygons_marginals)
for mm in range(len(found_polygons_text_region)): for mm in range(len(found_polygons_text_region)):
textregion = ET.SubElement(page, 'TextRegion') textregion = ET.SubElement(page, 'TextRegion')
textregion.set('id', 'r%s' % id_indexer) textregion.set('id', 'r%s' % id_indexer)
@ -1282,9 +1269,9 @@ class eynollah:
points_co += ',' points_co += ','
points_co += str(int((all_found_texline_polygons[mm][j][l][1] + page_coord[0]) / self.scale_y)) points_co += str(int((all_found_texline_polygons[mm][j][l][1] + page_coord[0]) / self.scale_y))
else: else:
points_co = points_co + str(int((all_found_texline_polygons[mm][j][l][0][0] + page_coord[2]) / self.scale_x)) points_co += str(int((all_found_texline_polygons[mm][j][l][0][0] + page_coord[2]) / self.scale_x))
points_co = points_co + ',' points_co += ','
points_co = points_co + str(int((all_found_texline_polygons[mm][j][l][0][1] + page_coord[0]) / self.scale_y)) points_co += str(int((all_found_texline_polygons[mm][j][l][0][1] + page_coord[0]) / self.scale_y))
elif curved_line and abs(slopes[mm]) > 45: elif curved_line and abs(slopes[mm]) > 45:
if len(all_found_texline_polygons[mm][j][l]) == 2: if len(all_found_texline_polygons[mm][j][l]) == 2:
points_co += str(int((all_found_texline_polygons[mm][j][l][0] + all_box_coord[mm][2] + page_coord[2]) / self.scale_x)) points_co += str(int((all_found_texline_polygons[mm][j][l][0] + all_box_coord[mm][2] + page_coord[2]) / self.scale_x))
@ -1298,7 +1285,6 @@ class eynollah:
if l < len(all_found_texline_polygons[mm][j]) - 1: if l < len(all_found_texline_polygons[mm][j]) - 1:
points_co += ' ' points_co += ' '
coord.set('points', points_co) coord.set('points', points_co)
add_textequiv(textregion) add_textequiv(textregion)
for mm in range(len(found_polygons_marginals)): for mm in range(len(found_polygons_marginals)):
@ -2002,12 +1988,13 @@ class eynollah:
text_regions_p = text_regions_p_1[:, :] # long_short_region[:,:]#self.get_regions_from_2_models(image_page) text_regions_p = text_regions_p_1[:, :] # long_short_region[:,:]#self.get_regions_from_2_models(image_page)
text_regions_p = np.array(text_regions_p) text_regions_p = np.array(text_regions_p)
if num_col_classifier == 1 or num_col_classifier == 2: if num_col_classifier in (1, 2):
try: try:
regions_without_seperators = (text_regions_p[:, :] == 1) * 1 regions_without_seperators = (text_regions_p[:, :] == 1) * 1
regions_without_seperators = regions_without_seperators.astype(np.uint8) regions_without_seperators = regions_without_seperators.astype(np.uint8)
text_regions_p = get_marginals(rotate_image(regions_without_seperators, slope_deskew), text_regions_p, num_col_classifier, slope_deskew, kernel=self.kernel) text_regions_p = get_marginals(rotate_image(regions_without_seperators, slope_deskew), text_regions_p, num_col_classifier, slope_deskew, kernel=self.kernel)
except: except Exception as e:
self.logger.error("exception %s", e)
pass pass
if self.plotter: if self.plotter: