qurator-spk/eynollah
1
0
Fork 0
mirror of https://github.com/qurator-spk/eynollah.git synced 2025-10-06 22:50:14 +02:00
Code Issues Projects Releases Wiki Activity

writing page contour correctly in xml output + ignore unsupported file types when loading images

Browse source
This commit is contained in:
vahidrezanezhad 2025-09-19 18:06:18 +02:00 • committed by kba
parent 994bc8a1c0
commit b38331b4ab
1 changed files with 4 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

6
src/eynollah/eynollah.py
View file

@ -1617,7 +1617,7 @@ class Eynollah:
else: else:
box = [0, 0, img.shape[1], img.shape[0]] box = [0, 0, img.shape[1], img.shape[0]]
cropped_page, page_coord = crop_image_inside_box(box, self.image) cropped_page, page_coord = crop_image_inside_box(box, self.image)
cont_page = cnt cont_page = [cnt]
#cont_page.append(np.array([[page_coord[2], page_coord[0]], #cont_page.append(np.array([[page_coord[2], page_coord[0]],
#[page_coord[3], page_coord[0]], #[page_coord[3], page_coord[0]],
#[page_coord[3], page_coord[1]], #[page_coord[3], page_coord[1]],
@ -3067,7 +3067,7 @@ class Eynollah:
if not self.ignore_page_extraction: if not self.ignore_page_extraction:
mask_page = np.zeros((text_regions_p_1.shape[0], text_regions_p_1.shape[1])).astype(np.int8) mask_page = np.zeros((text_regions_p_1.shape[0], text_regions_p_1.shape[1])).astype(np.int8)
mask_page = cv2.fillPoly(mask_page, pts=[cont_page], color=(1,)) mask_page = cv2.fillPoly(mask_page, pts=[cont_page[0]], color=(1,))
text_regions_p_1[mask_page==0] = 0 text_regions_p_1[mask_page==0] = 0
textline_mask_tot_ea[mask_page==0] = 0 textline_mask_tot_ea[mask_page==0] = 0
@ -4526,6 +4526,7 @@ class Eynollah:
if dir_in: if dir_in:
self.ls_imgs = os.listdir(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')]
elif image_filename: elif image_filename:
self.ls_imgs = [image_filename] self.ls_imgs = [image_filename]
else: else:
@ -5265,6 +5266,7 @@ class Eynollah_ocr:
def run(self, overwrite : bool = False): def run(self, overwrite : bool = False):
if self.dir_in: if self.dir_in:
ls_imgs = os.listdir(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')]
else: else:
ls_imgs = [self.image_filename] ls_imgs = [self.image_filename]