Someone has done training for the OCR-B font -- are you using that training data? Search the archives and you'll find the files to download. --Sven
On Wed, May 9, 2012 at 3:21 AM, [email protected] <[email protected]> wrote: > Hello, > I 'm trying to build up a program to read payment slip. The slip > contain a single OCR-B line on the bottom. That line is a record > containing some coded information. It is divided in four blocks > ( fields) but only three of them are delimited by special character > separator and contains a modulo10 control character. > In my first try I simply pass a Grayscale cropped image to Tesseract > and set in initilization 0123456789>+ as set of allowed char. > This gave me 70% of accuracy. > I know that for numbers I cannot apply a dictionary (that can boost > accuracy to near 99%) but for payments, 70 % of accuracy is not an > acceptable result. > Does anybody know how could I improve accuracy? > > thank you very much in advance > Franco > > -- > 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
