Thanks again. you may get better results using appropriate language data rather than just > the ascii range. Are the client documents sorted by language? >
I'm not sure how they have them organised, I just know they want an "automatic" solution... > > I am attaching files used - i had just copied some tables of ascii range - > you can delete symbols, add multiple copies of letters that are needed. > >> >> I'm still getting up and running with training (I'm doing it on linux as there appear to be more tools available that way). But I saw this comment from zdenop https://groups.google.com/forum/#!searchin/tesseract-ocr/train$20hall$20of$20fame/tesseract-ocr/tq2aHxxndpM/u5ldKIwUANIJ and it leads me to believe that getting much better trained data using the common fonts (arial, georgia, segoe, garamond) will not be any better then what is available? I have complete control over the image data I send to tesseract, so I don't care about skewing, exposure, etc, as my glyphs will always be straight, clear, and separated. For instance, I want to train for the ligatures ff, ffi, and ffl, which are not in the english or ascii ones, and are missing from even the common fonts like arial, but that my client files may contain. Should I train new eng or asc traineddata, or just create a new one for a smaller set of glyphs like these? Thanks again for your help. -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/368ecab0-4259-4d31-8b91-00ec82e1957d%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
