News for hackers. Stuff that matters.
Re:OCR by uploading images to Google Docs by Estragon on Friday November 26, 2010 @10:30AM (#1942557)
Incidentally, I noticed that if you try to use tesseract on an image taken from a Google Books page, you get terrible OCR accuracy. Anyone know why that is?
Re:OCR by uploading images to Google Docs by zzleeper on Friday November 26, 2010 @11:18AM (#1942638)
I recall that on some google-scanned books, there was some metadata from abbyy finereader. So that may be why.

Also, tesseract often needs to be configured.