Having experienced poor results using a couple of $400 software applications for Arabic optical character recognition (including the one offerred in an earlier question), I wondered if anyone could suggest something that actually works in practice, without a lot of manual intervention to correct for errors. Ideally, the software would have automatic cleanup or image enhancement (our data isn't always pristine), access to an API, and provide support to extended languages, like Farsi OCR, Dari OCR, Pashto OCR, Urdu OCR etc.
No comments:
Post a Comment