[nfbcs] OCR on Linux

[Jack Heim]

I used sane and tesseract to scann in the entire Dungeons & Dragons 
fifth edition players guide. It was a huge job and it is probably some 
of the toughest OCR there is.  It worked really well though. It took so 
long that by the time I was done I had worked out the best sane and 
tesseract settings. But I was getting really great accuracy toward the 
end. If I was scanning in a novel it probably would have been a breeze.




On 03/13/2017 02:04 PM, Mike Gorse via nfbcs wrote:
> Hi all,
>
> I just noticed that there's a CLI version of ABBYY FIneReader available
> for Linux. http://www.ocr4linux.com/en:start
>
> I should probably just test it, since it's possible to test it for free,
> but does anyone know whether it's worth buying / how it compares to
> open-source OCR options? I haven't ever seriously tried to OCR anything
> on Linux, but it might be nice to have the option.
>
> Thanks,
> -Mike
>
> _______________________________________________
> nfbcs mailing list
> nfbcs at nfbnet.org
> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
> To unsubscribe, change your list options or get your account info for
> nfbcs:
> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/john%40johnheim.com