Thanks as always. I guess will have to figure it out as I go along.Will surely share any learnings if I am able to crack this on my own.Talking of which your suggestion on the image scaling did help. I am still tweaking it a little bit as all my images are not of the same dimensions.(avg : 400x200 pixels).By scaling up to 2x I am able to get some better results.
Regards, Amrit. On Apr 12, 11:57 pm, Dmitri Silaev <[email protected]> wrote: > Well, this as interesting question for me too. > > A high-level interface (like "ResultIterator") for getting best > results at word level does not exist. You need to dive into the guts > of Tesseract. I can't give a ready recipe for this, only I can give > you some kind of a probable guiding thread: > > - Look into Wordrec::cc_recog() > - Look into Dict::DebugWordChoices() > - Check Dict::best_choices_ member > > It'd be great if share your findings with the community when you > complete your investigation. > > Warm regards, > Dmitri Silaev > > > > > > > > On Tue, Apr 12, 2011 at 10:55 PM, Amrit <[email protected]> wrote: > > Thanks Dmitri, > > > As a follow up, is there any way this can be done at a word level.As > > in if is a way I can get the top three most likely decoding results > > from tesseract for a given text image(individual words) > > > Regards, > > Amrit. > > > On Apr 12, 12:45 am, Dmitri Silaev <[email protected]> wrote: > >> One more subtlety: > > >> // This ensures Tesseract's "blob_choices" structures are filled > >> SetVariable("save_best_choices", "T"); > > >> On Tue, Apr 12, 2011 at 12:36 AM, Amrit <[email protected]> > >> wrote: > >> > Hi All, > >> > Is there a way of obtaining a N-best result list from > >> > tesseract output? Something similar to generating the decoding > >> > hypothesis.I was looking for a way to examine the results from the > >> > tesseract word decoding on my earlier posted task of identifying > >> > postal codes from a raw image. > >> > Please do let me know if anyone has come across a similar > >> > implementation.Any help is highly appreciated. > > >> > Regards, > >> > Amrit. > > >> > -- > >> > 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 > >> > athttp://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 > > athttp://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.
