← Back to team overview

sikuli-driver team mailing list archive

Re: [Question #146748]: need a tip to better get text from a region by OCR

 

Question #146748 on Sikuli changed:
https://answers.launchpad.net/sikuli/+question/146748

RaiMan posted a new comment:
I think in the moment, you have to live with it, when using the feature
Region.text().

I think your example is rather "high quality", since the relevant
information is read with a rather small font - "only" the structure of
the text is not recognized :-(

If you want to have better results currently, you either have to go into
the code details and try to optimize it in the openCV (image
preparation) and/or the tesseract area or use your own implementation
(mid last year I had a solution running with ImageMagick and tesseract -
as far as I remember with better results in those single line
situations).

-- 
You received this question notification because you are a member of
Sikuli Drivers, which is an answer contact for Sikuli.