I am trying to process the scanned text. What I would like to achieve is to split a page horizontally _between_ two lines of text, and then add some space around the croped part of the page. Any idea how to recognize the block of the text to avoid cutting the textlines?
A plethora of command-line scripts that perform geometric transforms, blurs, sharpens, edging, noise removal, and color manipulations.
2 posts • Page 1 of 1
- Posts: 23062
- Joined: 2007-07-02T17:14:51-07:00
- Authentication code: 1152
- Location: Sunnyvale, California, USA
Average the image down to one column and look for white values that represent spaces between lines. For this to work, the image cannot be rotated. All lines of text must be horizontal and the space between lines must avoid letters that drop down or rise up into the spaces.