If the coordinates of the rectangles are known, you can crop, and rescale to 300DPI if necessary, then send to Tesseract for recognition.
On Wednesday, August 6, 2014 10:11:32 AM UTC-5, Natan Katz wrote: > > Paul > > Thanks for your answer. Most of the pictures are of this form namely, when > there is a small amount of text in different areas of the picture, rather a > certain area. > > Natan > > בתאריך יום רביעי, 6 באוגוסט 2014 17:25:59 UTC+3, מאת Paul: >> >> You should post an example screenshot, so we have something to discuss. >> >> Paul >> >> Am Mittwoch, 6. August 2014 14:33:33 UTC+2 schrieb Natan Katz: >>> >>> >>> Hello >>> >>> I am trying to extract content of screenshots of applications and I get >>> very poor results since(I suspect) the words are scattered in the >>> pictures. Is there any way to improve this? >>> >>> Natan >>> >> -- 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 http://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/f66e3062-f3e4-42d9-af10-194b8f065c36%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
