Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Any relationship between optical character recognition and "deblurring" images into characters or words that the system recognizes?


OCR algorithms typically aren’t targeting heavily blurred text. It’s more about handling all the ways letters can look, including ligatures, determining paragraph breaks, detecting tables, ignoring staples and coffee stains, etc. than about correcting for bad scans/photos.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: