Base on my experience: it is very difficult to train on scaned images that you do not "create" (e.g. where you have no possibility to adjusts spaces as described in http://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract3#Generate_Training_Images )
In your case I would try to find similar font as on image and than I would create training images with enough spaces to avoid ""FAILURE! box overlaps". Zd. On Tue, Sep 28, 2010 at 7:43 PM, Bumbi <[email protected]> wrote: > And what can I do? > Otherwise if I resize the oroginal training image with Lanczos to 200% > it works. If I resize with Mitchell to 200% it get an error message. > > On szept. 28, 18:43, "Jimmy O'Regan" <[email protected]> wrote: > > On 28 September 2010 17:13, zdenko podobny <[email protected]> wrote: > > > > > send also box file for that image. > > > > No need. There's a known issue regarding spacing in training images, > > and given how small the characters are and how close they are to each > > other, I'm 99% sure that's what's happening here. > > > > -- > > <Leftmost> jimregan, that's because deep inside you, you are evil. > > <Leftmost> Also not-so-deep inside 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]<tesseract-ocr%[email protected]> > . > For more options, visit this group at > http://groups.google.com/group/tesseract-ocr?hl=en. > > -- 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.
