qurator-spk/eynollah
1
0
Fork 0
mirror of https://github.com/qurator-spk/eynollah.git synced 2025-09-01 13:29:58 +02:00
Code Issues Projects Releases Wiki Activity

ocr model renamed - image text font for ocr result is now using Charis-7.000 font (downloaded from here https://software.sil.org/charis/download/)

Browse source
This commit is contained in:
vahidrezanezhad 2025-07-16 14:00:12 +02:00
parent e54ebaa23e
commit e0f4a007e4
1 changed files with 78 additions and 70 deletions
Download patch file Download diff file Expand all files Collapse all files

20
src/eynollah/eynollah.py
View file

@ -318,7 +318,7 @@ class Eynollah:
if self.ocr and self.tr:
self.model_ocr_dir = dir_models + "/trocr_model_ens_of_3_checkpoints_201124"
elif self.ocr and not self.tr:
self.model_ocr_dir = dir_models + "/model_step_750000_ocr"#"/model_step_125000_ocr"#"/model_step_25000_ocr"#"/model_step_1050000_ocr"#"/model_0_ocr_cnnrnn"#"/model_23_ocr_cnnrnn"
self.model_ocr_dir = dir_models + "/model_eynollah_ocr_cnnrnn_20250716"
if self.tables:
if self.light_version:
self.model_table_dir = dir_models + "/modelens_table_0t4_201124"
@ -5129,7 +5129,7 @@ class Eynollah_ocr:
self.b_s = int(batch_size)
else:
self.model_ocr_dir = dir_models + "/model_step_900000_ocr"#"/model_step_25000_ocr"#"/model_step_1050000_ocr"#"/model_0_ocr_cnnrnn"#"/model_23_ocr_cnnrnn"
self.model_ocr_dir = dir_models + "/model_eynollah_ocr_cnnrnn_20250716"
model_ocr = load_model(self.model_ocr_dir , compile=False)
self.prediction_model = tf.keras.models.Model(
@ -5276,7 +5276,7 @@ class Eynollah_ocr:
if self.draw_texts_on_image:
font_path = "NotoSans-Regular.ttf" # Make sure this file exists!
font_path = "Charis-7.000/Charis-Regular.ttf" # Make sure this file exists!
font = ImageFont.truetype(font_path, 40)
for indexer_text, bb_ind in enumerate(total_bb_coordinates):
@ -5340,8 +5340,8 @@ class Eynollah_ocr:
tree1.write(out_file_ocr,xml_declaration=True,method='xml',encoding="utf8",default_namespace=None)
#print("Job done in %.1fs", time.time() - t0)
else:
max_len = 512
padding_token = 299
max_len = 512#280#512
padding_token = 299#1500#299
image_width = 512#max_len * 4
image_height = 32
@ -5436,6 +5436,11 @@ class Eynollah_ocr:
mask_poly = mask_poly[y:y+h, x:x+w, :]
img_crop = img_poly_on_img[y:y+h, x:x+w, :]
if self.export_textline_images_and_text:
if not self.do_not_mask_with_textline_contour:
img_crop[mask_poly==0] = 255
else:
#print(file_name, angle_degrees,w*h , mask_poly[:,:,0].sum(), mask_poly[:,:,0].sum() /float(w*h) , 'didi')
if not self.do_not_mask_with_textline_contour:
if angle_degrees > 3:
@ -5541,6 +5546,9 @@ class Eynollah_ocr:
cropped_lines_bin.append(img_fin)
if self.export_textline_images_and_text:
if img_crop.shape[0]==0 or img_crop.shape[1]==0:
pass
else:
if child_textlines.tag.endswith("TextEquiv"):
for cheild_text in child_textlines:
if cheild_text.tag.endswith("Unicode"):
@ -5727,7 +5735,7 @@ class Eynollah_ocr:
if self.draw_texts_on_image:
font_path = "NotoSans-Regular.ttf" # Make sure this file exists!
font_path = "Charis-7.000/Charis-Regular.ttf" # Make sure this file exists!
font = ImageFont.truetype(font_path, 40)
for indexer_text, bb_ind in enumerate(total_bb_coordinates):