qurator-spk/eynollah
1
0
Fork 0
mirror of https://github.com/qurator-spk/eynollah.git synced 2025-10-26 23:34:13 +01:00
Code Issues Projects Releases Wiki Activity

training/gt_gen_utils: fix type errors, comment out dead code

Browse source
This commit is contained in:
kba 2025-10-17 14:21:05 +02:00
parent af74890b2e
commit 557fb227f3
1 changed files with 14 additions and 15 deletions
Download patch file Download diff file Expand all files Collapse all files

29
src/eynollah/training/gt_gen_utils.py
View file

@ -252,6 +252,7 @@ def get_textline_contours_for_visualization(xml_file):
x_len, y_len = 0, 0
for jj in root1.iter(link+'Page'): for jj in root1.iter(link+'Page'):
y_len=int(jj.attrib['imageHeight']) y_len=int(jj.attrib['imageHeight'])
x_len=int(jj.attrib['imageWidth']) x_len=int(jj.attrib['imageWidth'])
@ -293,6 +294,7 @@ def get_textline_contours_and_ocr_text(xml_file):
x_len, y_len = 0, 0
for jj in root1.iter(link+'Page'): for jj in root1.iter(link+'Page'):
y_len=int(jj.attrib['imageHeight']) y_len=int(jj.attrib['imageHeight'])
x_len=int(jj.attrib['imageWidth']) x_len=int(jj.attrib['imageWidth'])
@ -362,7 +364,7 @@ def get_layout_contours_for_visualization(xml_file):
link=alltags[0].split('}')[0]+'}' link=alltags[0].split('}')[0]+'}'
x_len, y_len = 0, 0
for jj in root1.iter(link+'Page'): for jj in root1.iter(link+'Page'):
y_len=int(jj.attrib['imageHeight']) y_len=int(jj.attrib['imageHeight'])
x_len=int(jj.attrib['imageWidth']) x_len=int(jj.attrib['imageWidth'])
@ -637,7 +639,7 @@ def get_images_of_ground_truth(gt_list, dir_in, output_dir, output_type, config_
link=alltags[0].split('}')[0]+'}' link=alltags[0].split('}')[0]+'}'
x_len, y_len = 0, 0
for jj in root1.iter(link+'Page'): for jj in root1.iter(link+'Page'):
y_len=int(jj.attrib['imageHeight']) y_len=int(jj.attrib['imageHeight'])
x_len=int(jj.attrib['imageWidth']) x_len=int(jj.attrib['imageWidth'])
@ -645,15 +647,12 @@ def get_images_of_ground_truth(gt_list, dir_in, output_dir, output_type, config_
if 'columns_width' in list(config_params.keys()): if 'columns_width' in list(config_params.keys()):
columns_width_dict = config_params['columns_width'] columns_width_dict = config_params['columns_width']
metadata_element = root1.find(link+'Metadata') metadata_element = root1.find(link+'Metadata')
comment_is_sub_element = False num_col = None
for child in metadata_element: for child in metadata_element:
tag2 = child.tag tag2 = child.tag
if tag2.endswith('}Comments') or tag2.endswith('}comments'): if tag2.endswith('}Comments') or tag2.endswith('}comments'):
text_comments = child.text text_comments = child.text
num_col = int(text_comments.split('num_col')[1]) num_col = int(text_comments.split('num_col')[1])
comment_is_sub_element = True
if not comment_is_sub_element:
num_col = None
if num_col: if num_col:
x_new = columns_width_dict[str(num_col)] x_new = columns_width_dict[str(num_col)]
@ -1739,15 +1738,15 @@ tot_region_ref,x_len, y_len,index_tot_regions, img_poly
def bounding_box(cnt,color, corr_order_index ): # def bounding_box(cnt,color, corr_order_index ):
x, y, w, h = cv2.boundingRect(cnt) # x, y, w, h = cv2.boundingRect(cnt)
x = int(x*scale_w) # x = int(x*scale_w)
y = int(y*scale_h) # y = int(y*scale_h)
#
w = int(w*scale_w) # w = int(w*scale_w)
h = int(h*scale_h) # h = int(h*scale_h)
#
return [x,y,w,h,int(color), int(corr_order_index)+1] # return [x,y,w,h,int(color), int(corr_order_index)+1]
def resize_image(seg_in,input_height,input_width): def resize_image(seg_in,input_height,input_width):
return cv2.resize(seg_in,(input_width,input_height),interpolation=cv2.INTER_NEAREST) return cv2.resize(seg_in,(input_width,input_height),interpolation=cv2.INTER_NEAREST)