samedi 27 juin 2015

Best way to scan images like this in tesseract? which PIM?

So how should I parametrize tesseract to best scan images like

The worksheet type of worksheet I want to scan (borrow from Google Image search)

Is there anyway I can improve tesseract accuracy in parsing this? Or have it realize that it should read vertically. I am using pytesseract to read in the file? Can I modify the pim with pytesseract? Is there another python binding I should be using? Any help would be appreciated.

Aucun commentaire:

Enregistrer un commentaire