← Back to team overview

cuneiform team mailing list archive

Re: API issues

 

Jussi,

Thank you. After some experimentation it appears that cuneiform can do
two characters, so I was able to get the desired effect by passing to it
a doubled-up image (i.e. "AA" instead of "A").
So - it works now!

Thank you for your help and the fantastic job you are doing with porting
cuneiform!

Igor



On Thu, 2010-08-19 at 10:01 +0300, Jussi Pakkanen wrote:
> On Wed, Aug 18, 2010 at 8:27 PM, Igor Filippov
> <igor.v.filippov@xxxxxxxxx> wrote:
> 
> > I still cannot get cuneiform to recognize a single stand-alone character
> > though.
> 
> At least Tesseract requires a line (or possibly several) of text
> before it can do detection. This is because it needs statistics for
> several blobs together before it can make a reliable classification.
> I'm assuming the same holds for Cuneiform.
> 
> _______________________________________________
> Mailing list: https://launchpad.net/~cuneiform
> Post to     : cuneiform@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~cuneiform
> More help   : https://help.launchpad.net/ListHelp





References