Thanks Calomer. Bonny, is the language you're trying to improve using a different set of characters (alphabet)? If so, you'll need to do a lot of training as Calomer described. Otherwise you'll just need some tweaks. The font may be an issue. --Sven
On Thu, Sep 29, 2011 at 12:39 PM, Calomer <[email protected]> wrote: > I'll try my best to answer, tho I'm hardly eligible. > > According to training instructions (on > http://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract3) > and general OCR knowledge, you cannot train solely by new characters. > You need training images, you need to create boxes (with any box > editor, but I only used Qt Box Editor). Once you create new boxes > around your new tiff image, and label them accordingly, you should be > ready for training. > > Keep in mind, you'll need at least 12 low x-height in pixels > (preferably around 20 pixels), variety in images would be nice for > increased performance. > > Follow training instructions, train your own language file, try OCR > again, if you fail again, I'm sure someone else who has wider > knowledge than me should be able to answer your further questions. > > On Sep 29, 2:44 pm, Bonny <[email protected]> wrote: >> Nobody know or the question is too silly? > > -- > 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 > -- ``All that is gold does not glitter, not all those who wander are lost; the old that is strong does not wither, deep roots are not reached by the frost. >From the ashes a fire shall be woken, a light from the shadows shall spring; renewed shall be blade that was broken, the crownless again shall be king.” -- 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
