Skip to content

Latest commit

 

History

History
49 lines (33 loc) · 1.95 KB

README.md

File metadata and controls

49 lines (33 loc) · 1.95 KB
Raw

ElmalzmaWatermarkRemover

ElmalzmaWatermarkRemover is a python script that removes the annoying watermark that is typically found at your studying sheets using OpenCV

Running the script (basic instructions)

  • you need to install python 3
  • run pip install -r requirements.txt to install required python packages

Usage

python filterwm.py -p | -f | -s target -t threshold -i iterations

Options:
     -p              Process images in pdf file named `target`
     -f              Process images in folder named `target`
     -s              Process single image file named `target`
     -t threshold    Thershold level [0-255] to apply binarization on at the watermark area
     -i iterations   Number of iterations in the watermark detection using Opening/Closing  Morphological process

Examples

python filterwm.py -s 1.jpg -t 53 -i 9

python filterwm.py -s 2.jpg -t 53 -i 15

Limitions

  • Thershold level and number of iterations must be entered and they depend on the condition of the scanning. The best way to find the best values is trial and error. Typical values for thershold is in range 45 - 75 and 5 - 15 for number of iterations for good quality images

  • The script does not detect the edges of the paper. Please crop the images to include the paper only

  • The script may not work if the page is already copied, or the image is in grayscale or taken by a mobile phone in low light