My product manager just asked me for at OCR text recognition operation to be used with the hybrid application i've been developing. I have been looking at tesseract, Ocrad, and some other alternatives, yet they seem so inconsistent.
Is there any better alternatives for OCR?
I am using it for scraping text from business cards. I know this poses some problems when fonts, colors etc. are not identical. The inconsistency I am facing is most like due to this.