Yaroslav, Tesseract should work on images "out of the box" so to speak.
If accuracy is 0%, try looking at the size of the images you are feeding into tesseract. The wiki has specifics on what size images and what sized fonts are best to use for tesseract to recognize. On May 20, 7:52 pm, Yaroslav Bulatov <[email protected]> wrote: > I already have a system which segments characters and would like to > use tesseract to recognize individual characters, is it possible to > tell tesseract to skip the bounding-box finding process? > > I tried it with "tesseract crop.tif result nobatch myletters", where > myletters restricted whitelist to digits only, and I'm getting no > output for 84 of 100 crops. Remaining 16 were treated as multiline > images, and overall accuracy was 0 --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/tesseract-ocr?hl=en -~----------~----~----~----~------~----~------~--~---
