How to cleanup an image before OCR ?

Discussions about Tesseract OCR integration in GdPicture.
Post Reply
Posts: 1
Joined: Tue May 14, 2013 9:41 am

How to cleanup an image before OCR ?

Post by jbergas » Thu May 23, 2013 4:40 pm

I'm trying to do OCR with the image i've attached and the result is "25755" instead "421,22" (300dpi, converted to 1bpp, OCRTesseractSetPassCount=5, OCRTesseractSetOCRContext=0 Document , CharWhiteList="0123456789.,-" )

I know the problem is the "noise" at the upper area, how can i erase this "noise" ?

I've seen the demo "document clean up.exe", and i've tried to solve it with "Removelines","RemoveBlob","FxBitonalDespeckleMore" methods unsuccessful.

000010_ZonaOCRLeida.jpg (4.43 KiB) Viewed 2358 times

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests