Newbie here and OCR

89 views
Skip to first unread message

Felipe Carvalho

unread,
Nov 9, 2017, 5:12:50 AM11/9/17
to tesseract-ocr
Hey guys, really beginning to touch OCR so bare with me.

Is this already using image processing before sending it to tesseract? Like is it using liptonica? 

Also imagine i'm just searching for very specific locations, and some times I'm just looking for icons instead of text but mostly text. What would be the best mode? I'm talking about Page Segmentation Mode

I'm using a small image of clear text with 2 words and it's recognizing most of the characters correctly but the P's are H's and Awakening was recognized as Awakemng


Any tips, help is apreciated
Reply all
Reply to author
Forward
0 new messages