Hi: I have a little notebook filled with annotations in my handwriting. I would wish to transform its contents into a plain ASCII text file, so that I could search for information in it. For example, I know somewhere in the notebook I once wrote something about the vitamin B2 contents in milk produced by certain manufacturer. The problem is where exactly, on what page, the data are written (it's about a 50-page notebook).

First I would have to scan the notebook, getting say, a PDF file. And then use this file as input to an optical character recognition (OCR) program, asking the program to write an ASCII file as its output. Having never done such a thing before, my question is what is the OCR program I could use being as I am a linux user. I guess OCR programs work with files as input.

What's asked.

Strike that, I'll just merge your threads. Bottom of the page, the "Similar Threads" feature, second link "Optical Character Recognition in Slackware64" to start with.

OCR only works with typeset fonts, not handwriting. Handwriting is extremely hard to parse, still an active research topic in machine learning.

If you want to do OCR (w/ typeset fonts) then I would use gocr or tesseract. Both are in SlackBuilds.org

1 members found this post helpful.

Oh, I did not give the matter enough thinking. I new that printed writing could be recognized by a machine and I should have guessed that handwriting is quite another thing. Thanks a lot for the information, metageek.