From kde-devel  Sat Oct 17 08:10:24 2009
From: Brad Hards <bradh () frogmouth ! net>
Date: Sat, 17 Oct 2009 08:10:24 +0000
To: kde-devel
Subject: Re: Volunteering to port Kooka to KDE4
Message-Id: <200910171910.24497.bradh () frogmouth ! net>
X-MARC-Message: https://marc.info/?l=kde-devel&m=125576706904541

On Saturday 17 October 2009 07:10:18 John Layt wrote:
>  If you can 
> make the OCR function an embeddable widget, kpart and Kipi plugin, then it
>  can  be used by any image management app such as Gwenview on the images it
>  manages, and you can then write a new specialised scanning app from
>  scratch around it.
>
> With regard to OCR, I guess your main support target these days would be 
> tesseract, Google's OCR engine (http://code.google.com/p/tesseract-ocr/),
>  but  I think most of the old targets are still around.
I've also looked at OCR a little. Tesseract doesn't have any page layout 
analysis capabilities (e.g. it can't recognise columns). There are some tools 
in Ocropus, and they might be useful for other things (e.g. for cut-n-paste in 
Okular).

Also, note that Ocropus and Tesseract are Apache v2 licensed.

Brad
  
 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<