It said 3.0 would have Chinese. I guess it's delayed? If I am on my own how would I go about training for Chinese/Japanese?
Also these images I'm providing are pretty "black & white" no pun intended. How is OCR on regular real photos outdoor photos of signs and billboards etc? Google has patented an engine that can accurately extract text from images and video. How would their algorithms be different from regular OCR? Why can't OCR companies do the same thing? Here's what I mean. Take this photo for example. Why can't regular OCR engines extract this text? http://i35.tinypic.com/33l1et3.jpg Thanks On Sep 24, 6:23 am, SteveP <[email protected]> wrote: > 3.0 is available by getting the latest sources using svn. I do not > believe there is an official release package yet. I do not see > Chinese or Japanese yet. 3.0 should handle the Spanish sentence. > Before 3.0, tesseract performance was lower on small amounts of text. > There is a wiki with documentation in the same area with the source > distributions. It includes a doc about training. It is not updated > yet with all the new info for 3.0, though. For Spanish, I suspect it > would do fine without training. > > On Sep 23, 4:43 am, Barney <[email protected]> wrote: > > > I tried a very simple Spanish sentence today and the OCR spit out one > > very odd spelling. A bit concerning. > > > Input: > > [IMG]http://i36.tinypic.com/35i5chj.jpg[/IMG] > > > Output: > > [i]El ctoiic se estrena con inundaciones[/i] > > > Chinese is coming for version 3.0 right? Do you have an ETA? Since > > Japanese is derived from Chinese how would I train the engine for > > Japanese? Or any other languages for that matter. > > > Lastly, I just want to confirm that this is wholly open source. How > > does the licensing work? > > > Thank you --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
