Hi Filipe, Please search the archives -- micke and Dmitri Silaev had a conversation about this in April. http://groups.google.com/group/tesseract-ocr/search?group=tesseract-ocr&q=character-level+confidence&qt_g=Search+this+group
--Sven On Wed, Aug 17, 2011 at 2:17 PM, _Filipe <[email protected]> wrote: > Hello guys! > > I got a really tough work in text recognition area and I'm using > Tesserct-ocr as my OCR tool. > The problem consist in recognize IDs printed on steel slabs and > identify them. > Using only Tesseract I recognize no text. So a detection and > segmentation phase is necessary. > With that segmentation and after training the tesseract with our > dictionary, the recognition rate is about 60%. > > We find a way to increase it in a probabilistic way, that would change > characters, fixing common errors. > To accomplish that we will need the confidence returned by the ocr > tool in a character base way. > I saw in code the function which returns it from a word. > How could I get it for a character, is it possible with the current > API? > if not, is there a way to change the tesseract code to get that? Where > should I start from? > > Thanks in advance. > > Best Regards > > Filipe > > -- > 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
