Thanks for the reply Zdenko! It does seem that tweaking the segmentation (through the -psm flag) produces better results for the attached images.
Cheers! Nick On Saturday, December 29, 2012 2:12:25 PM UTC-8, zdenop wrote: > > try to use page segmentation mode. E.g. "Treat the image as a single word" > (or text line or uniform block of text) will produce results. > As far as I remember discussion on this forum tesseract is not suitable > for handwritten text... > > Zdenko > > > On Fri, Dec 28, 2012 at 11:55 PM, Nick Jalbert > <[email protected]<javascript:> > > wrote: > >> I'm trying to use Tesseract to detect the presence of text in images. >> I'm not concerned about the accuracy of the OCR, but would like Tesseract >> to return a non-empty answer if there is something in the picture. I have >> a few questions: >> >> 1. Are there any other image processing algorithms/tools I should look >> into besides Tesseract for this task? >> >> 2. Are there any config settings I might want to tweak to make Tesseract >> behave better as a binary classifier? I tried tweaking the "dictionary >> trust" settings as described in the FAQ, but didn't see much difference. >> >> 3. Attached are a few images that Tesseract 3.02.02 tells me are blank. >> Are these just pathological cases, or am I doing something wrong? >> >> Thanks for any help! >> Nick >> >> -- >> 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]<javascript:> >> To unsubscribe from this group, send email to >> [email protected] <javascript:> >> For more options, visit this group at >> http://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
