def run_seg(request, pk):
page = get_object_or_404(Page, pk=pk)
img_path = page.get_image_path()
raw_data = serializers.serialize("python",
Region.objects.filter(page_id=pk))
region_lst = [d['fields'] for d in raw_data]
char_lst = charseg(img_path, region_lst, pk)
for ch in char_lst:
char_id = '{0}L{1:02}R{2:02}C{3:03}'.format(pk, ch['line_no'],
ch['region_no'],
ch['char_no'])
character = Character(
id=char_id,
page_id=pk,
char=ch['char'],
left=ch['left'],
right=ch['right'],
top=ch['top'],
bottom=ch['bottom'],
line_no=ch['line_no'],
region_no=ch['region_no'],
char_no=ch['char_no'],
is_correct=-9 # has not chut char image
)
character.save()
cut_char.delay(pk) #offline task cut char image
return JsonResponse(char_lst, safe=False)
def segment_one_page(page_id, image_name, text):
if text == text1:
print 'equal----'
try:
image = io.imread(image_name, 0)
total_char_lst = process_page(image, text, page_id)
character_lst = []
for ch in total_char_lst:
character = Character(id=ch.char_id.strip(),
page_id=page_id,
char=ch.char,
image=ch.char_id.strip() + u'.jpg',
left=ch.left,
right=ch.right,
top=ch.top,
bottom=ch.bottom,
line_no=ch.line_no,
char_no=ch.char_no,
is_correct=False)
#character_lst.append(character)
print character
character.save()
#Character.objects.bulk_create(character_lst)
except:
print 'missing image file'
def test_file():
vol_no = u'01'
page_no = u'001'
output = subprocess.check_output(
'cd /media/DATA/work/image-preprocessing/binary_image/%s/%s/; ls *.txt 2>/dev/null'
% (vol_no, page_no),
shell=True)
pos = output.find('.txt')
if pos == -1:
sys.exit(-1)
page_id_prefix = output[:pos - 1]
page_id = page_id_prefix + 'a'
image_name = u'/media/DATA/work/image-preprocessing/binary_image/%s/%s/%s.jpg' % (
vol_no, page_no, page_id)
text_name = u'/media/DATA/work/image-preprocessing/binary_image/%s/%s/%s.txt' % (
vol_no, page_no, page_id)
image = io.imread(image_name, 0)
# image = io.imread(u'/home/xianbu/custom/1a.jpg', 0)
text = u''
with open(text_name, 'r') as f:
text = f.read().decode('utf-8')
text1 = text
process_page(image, text, page_id)
total_char_lst = process_page(image, text, page_id)
import django
django.setup()
from segmentation.models import Character
character_lst = []
for ch in total_char_lst:
character = Character(id=ch.char_id.strip(),
page_id=page_id,
char=ch.char,
image=ch.char_id.strip() + u'.jpg',
left=ch.left,
right=ch.right,
top=ch.top,
bottom=ch.bottom,
line_no=ch.line_no,
char_no=ch.char_no,
is_correct=False)
#character_lst.append(character)
print character
character.save()
def segment_one_page(page_id, image_name, text):
if text == text1:
print 'equal----'
try:
image = io.imread(image_name, 0)
total_char_lst = process_page(image, text, page_id)
character_lst = []
for ch in total_char_lst:
character = Character(id=ch.char_id.strip(), page_id=page_id, char=ch.char,
image=ch.char_id.strip() + u'.jpg',
left=ch.left, right=ch.right,
top=ch.top, bottom=ch.bottom,
line_no=ch.line_no, char_no=ch.char_no,
is_correct=False)
#character_lst.append(character)
print character
character.save()
#Character.objects.bulk_create(character_lst)
except:
print 'missing image file'
def test_file():
vol_no = u'01'
page_no = u'001'
output = subprocess.check_output(
'cd /media/DATA/work/image-preprocessing/binary_image/%s/%s/; ls *.txt 2>/dev/null' % (vol_no, page_no),
shell=True)
pos = output.find('.txt')
if pos == -1:
sys.exit(-1)
page_id_prefix = output[:pos - 1]
page_id = page_id_prefix + 'a'
image_name = u'/media/DATA/work/image-preprocessing/binary_image/%s/%s/%s.jpg' % (vol_no, page_no, page_id)
text_name = u'/media/DATA/work/image-preprocessing/binary_image/%s/%s/%s.txt' % (vol_no, page_no, page_id)
image = io.imread(image_name, 0)
# image = io.imread(u'/home/xianbu/custom/1a.jpg', 0)
text = u''
with open(text_name, 'r') as f:
text = f.read().decode('utf-8')
text1 = text
process_page(image, text, page_id)
total_char_lst = process_page(image, text, page_id)
import django
django.setup()
from segmentation.models import Character
character_lst = []
for ch in total_char_lst:
character = Character(id=ch.char_id.strip(), page_id=page_id, char=ch.char,
image=ch.char_id.strip() + u'.jpg',
left=ch.left, right=ch.right,
top=ch.top, bottom=ch.bottom,
line_no=ch.line_no, char_no=ch.char_no,
is_correct=False)
#character_lst.append(character)
print character
character.save()
def run_seg(request,pk):
page = get_object_or_404(Page, pk=pk)
img_path = page.get_image_path()
raw_data = serializers.serialize("python",Region.objects.filter(page_id=pk))
region_lst = [d['fields'] for d in raw_data]
char_lst = charseg(img_path,region_lst, pk)
for ch in char_lst:
char_id = '{0}L{1:02}R{2:02}C{3:03}'.format(pk,ch['line_no'],ch['region_no'],ch['char_no'])
character = Character(
id=char_id,
page_id= pk,
char=ch['char'],
left=ch['left'],
right=ch['right'],
top=ch['top'],
bottom=ch['bottom'],
line_no=ch['line_no'],
region_no = ch['region_no'],
char_no=ch['char_no'],
is_correct=-9 # has not chut char image
)
character.save()
cut_char.delay(pk)#offline task cut char image
return JsonResponse(char_lst, safe=False)
