On 28 July 2010 07:46, caro <[email protected]> wrote: > I process the training using 32 images per caracter, from two > different videos. > I only have digits from 0 to 9 to recognize. > > Sometimes, it does not recognize some digits in the video. The OCR > gives nothing. > I do not understand how it is possible that I do not have anything. I > imagine it computes probabilities for each digit to be recognize, and > give the best as output. > > So first question: is it normal that sometimes, I have no output from > the OCR (while it will recognize nearly the same digit at the next > frame)?
I can't really say, without seeing the images. If the frame is blurred - i.e., if tesseract sees nothing that it's reasonably confident is a character - then yes, this is what you should expect. > Images are not different, and the digits are quite similar. So it is a > bit strange that the OCR does not seem stable. > > Second question: is it possible to keep the 3 best caracters > recognized? I really have no idea what you meant to ask here. > Then if one is wrong in the word, we can correct it using a priori > information, and maybe in taking the 2 best caracter found by the OCR? > > Thank you for your help, > Caroline > > > -- > 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. > > -- <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]. For more options, visit this group at http://groups.google.com/group/tesseract-ocr?hl=en.
