You might need more samples. The training process usually uses a minimum of 5-10 samples of each character in each font.Did you get any errors from applybox? See Important under Run Tesseract for Training. Ray.
On Tue, Mar 24, 2009 at 2:32 AM, bergheil <[email protected]>wrote: > > Hello, nobody can please explain me what is wrong in my training > process? > Please help me. > > On 20 Mar, 08:38, bergheil <[email protected]> wrote: > > Ciao a tutti, > > I'm building a training for recogninze the cmc7 fonts (only numbers > > and chars /^>! ). > > I have used for training an openoffice file with 3 lines of fonts > > wrote by myself and 8 file with the real document to recognize (bank > > check). After the training tesseract recognize the whole openoffice > > file and 50% of the 8 bank check. > > Yesterday I have added 21 new bank check, and there'isnt no > > improvement!!!! That's is impossible, so I can image that there is > > something wrong in my training works. > > The training documentation says : "The first step is to determine the > > full character set to be used, and prepare a text or word processor > > file containing a set of examples." , I have used the real document > > for the training not the word processor file (just 3 lines really), is > > it wrong? > > Please help me. > > Regards > > > > O.S.: debian 4.0 > > Tesseract: 2.03 > > > --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
