I generally do image resizing to help me to correct errors like this. For ex, for your test1.bmp, I did:
*convert test1.bmp -resize 400% testnew.bmp* I used imagemagick to resize the image. After this, tesseract identified ':' correctly. Though sometimes, image resizing introduces some other errors like detection of non existing spaces, which I am still trying to figure out how to avoid. On Tuesday, June 7, 2016 at 12:28:45 AM UTC+5:30, Carlo wrote: > > Hi All, > > I am trying to perform an OCR recognition and I am noticing a strange > behaviour: with image "test1.bmp" (in attach) the engine returns: > > DOP Z IT001E34343434 (with a "Z" instead of ":") > > With image "test2.bmp" (in attach), that is a portion wider then > test1.bmp, the engine returns: > > maggiore > DOP : IT001E34343434 (correct!) > > Why, with the first image, the engine detects a "Z" instead of ":" and in > the other case the engine detects correctly a ":"? > > Many thanks. > Best Regards. > > Carlo > -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/2c33e06e-3549-432e-bc68-cbd25e00a0c0%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
