Skip to content

Commit

Permalink
Add unpaper and deskew
Browse files Browse the repository at this point in the history
  • Loading branch information
sbrunner committed Oct 31, 2018
1 parent 4026cfd commit bfec334
Show file tree
Hide file tree
Showing 2 changed files with 24 additions and 7 deletions.
8 changes: 7 additions & 1 deletion Dockerfile
Expand Up @@ -3,10 +3,16 @@ FROM ubuntu:cosmic
RUN \
. /etc/os-release && \
apt-get update && \
apt-get install --assume-yes --no-install-recommends python3-yaml graphicsmagick scantailor pdftk-java tesseract-ocr tesseract-ocr-fra libimage-exiftool-perl && \
apt-get install --assume-yes --no-install-recommends python3-yaml graphicsmagick scantailor pdftk-java tesseract-ocr tesseract-ocr-fra libimage-exiftool-perl unpaper unzip curl && \
apt-get clean && \
rm --recursive --force /var/lib/apt/lists/*

RUN \
curl http://galfar.vevb.net/store/deskew-125.zip > /tmp/deskew-125.zip && \
unzip /tmp/deskew-125.zip -d /opt && \
chmod +x /opt/Deskew/Bin/deskew &&\
rm /tmp/deskew-125.zip

CMD ["/opt/scan"]

VOLUME /source \
Expand Down
23 changes: 17 additions & 6 deletions scan
Expand Up @@ -50,6 +50,10 @@ def transform(config, root_folder):
os.makedirs(os.path.join(root_folder, 'process'))
if not os.path.exists(os.path.join(root_folder, 'scantailor')):
os.makedirs(os.path.join(root_folder, 'scantailor'))
if not os.path.exists(os.path.join(root_folder, 'unpaper')):
os.makedirs(os.path.join(root_folder, 'unpaper'))
if not os.path.exists(os.path.join(root_folder, 'deskew')):
os.makedirs(os.path.join(root_folder, 'deskew'))
for img in config['images']:
shutil.copyfile(os.path.join(root_folder, img), os.path.join(root_folder, 'process', os.path.basename(img)))
img = os.path.join(root_folder, 'process', os.path.basename(img))
Expand All @@ -68,12 +72,6 @@ def transform(config, root_folder):
maring_vertical=5
if not config['args']['no_crop']:
if not config['args']['new_crop']:
call([
'scantailor-cli', '--dpi=300', '--content-detection=normal',
'--content-box=0x0:1000x1000',
'--output-dpi=300', '--color-mode=color_grayscale', '--white-margins=true',
'--margins-left=0', '--margins-right=0', '--margins-top=0', '--margins-bottom=0',
img, os.path.join(root_folder, 'scantailor')])
old_crop(img, round(marging_horizontal/10/2.51*300), round(maring_vertical/10/2.51*300))
"""
call(
Expand Down Expand Up @@ -121,6 +119,19 @@ def transform(config, root_folder):
print("Ignore image with no content: {}".format(img))
continue

call([
'scantailor-cli', '--dpi=300', '--content-detection=normal',
'--content-box=0x0:1000x1000',
'--output-dpi=300', '--color-mode=color_grayscale', '--white-margins=true',
'--margins-left=0', '--margins-right=0', '--margins-top=0', '--margins-bottom=0',
img, os.path.join(root_folder, 'scantailor')
])
call([
'unpaper', img, os.path.join(root_folder, 'unpaper', os.path.basename(img))
])
call([
'/opt/deskew/Bin/deskew', '-o', os.path.join(root_folder, 'deskew', os.path.basename(img)), img
])
img2 = os.path.join(root_folder, os.path.basename(img))
subprocess.call(convert + [img, img2])
images.append(img2)
Expand Down

0 comments on commit bfec334

Please sign in to comment.