Re: [orca-list] OCR GUI?

[hackingKK <hackingkk gmail com>]

On Friday 12 February 2010 02:04 PM, Jacob Schmude wrote:
> Hi everyone
> Well, I've had another look at Linux OCR and I must say I'm very
> impressed. Tesseract in particular is loads better than when I tried it
> over a year back, and imho good enough to use in almost all situations.
> It's not failed me yet. At the moment, however, I do all my scanning and
> OCR from the command-line. I don't mind this, but I'd like to set it up
> for a friend of mine and that person would prefer a GUI. I have tried
> gscan2pdf, but try as I might I cannot get it to produce any OCR output
> no matter what settings I tell it to use. I thought perhaps the text
> control where the OCR was displayed had some issues, but it turns out
> that even when the PDF was saved and converted to text there's nothing.
>    

Well, a few students whom I am mentoring are already into creating a 
kind of GUI upon kies_p2T.
We found ocropus pritty good and cuniform better as well.
Both of them give good results infact more than 90% accuracy.

We are building a pygtk based GUI which will give features similar to 
open book or other similar proprietory softwares.

Happy hacking.
Krishnakant.