chitrankan OCR - iscii and windows

[Shree Devi Kumar]

The discussion regarding SanskritOCR led me to try out chitrankan again as it directly converts the scanned image to devanagari and unicode.

 

I was trying to train the font to improve the recognition rate. the problem that I am facing is that the input box does not accept keyboard input (i am using XP).

 

The documentation suggests that chitrankan uses an in-built iscii editor for cut, paste etc and allows for inscript and phonetic keyboards for input.

 

Hence if I copy and paste a letter from the converted output then it works, but trying to directly type in the box does not.

 

Is there anything special that needs to be done to use ISCII under XP?

 

Thanks,

Shree

[Hariram]

Which Version of Chitrankan you trying? That from the Hindi Software CD of MIT?

 

If yes, it was 9 years old beta version. It is not wise to waste time in this.

Unconfirmed news that It's 2009 version is under development by CDAC,Kolkata with some collaborations.

--
हरिराम
प्रगत भारत <http://hariraama.blogspot.com>

[Suresh Shukla]

CDAC would benefit by adopting open source model. Community effect
observed in many successful open source projects can work wonders in
such projects - including early testing, fixes, porting to other
platforms, reviews, feedback etc.

I hope this message reaches them,
Suresh

2009/5/27 Hariram <hari...@gmail.com>:

[Shree Devi Kumar]

I downloaded the files from http://www.ildc.gov.in/htm/ocr_spell.htm

It is probably the old 'lite' version as the full fledged product is priced at Rs. 10,000 for single license.

Do not know whether CDAC will provide a free version with the new 2009 release.

Thanks for the update.

Shree