0

I try to make Unpaper to work, it's a very powerfull tool to post-process scanned book pages. I read the Doc but still can't make it work neither on Windows nor on Linux. My picture is a Bitmap I converted to .pbm format with Gimp (also tried pgm and pnm). Then I use this command:

unpaper -l double --pre-rotate -90 inputFile.pbm outputFile.pbm

(-l double means I have 2 pages in one sheet and --pre-rotate -90 turns the image to the left)

It should do something like this:

enter image description here

But my output file has no differences between the input, just rotated.

1 Answer 1

0

I finally found out the issue. The problem was the conversion to the .pbm format done by Gimp (v2.6.11). The converted file was not binary (only white or black pixels) whereas Wikipedia says "PBM is for bitmaps (black and white, no grays)". So I used ImageMagick for the conversion and now unpaper works perfectly fine and I have excellent results with tesseract 3.

1
  • 1
    What exactly did you do? Did you convert the input images into another format before processing them with unpaper? If so, what format did you convert them to?
    – zrajm
    Sep 30, 2017 at 10:57

You must log in to answer this question.