Scanning Text - OCR

Its quite difficult to find an open source OCR (Optical Character Recognition) program. I tried Kooka (Ocrad) and also gOCR and in both cases I did not get very good results, I may not have been using the programs properly, but I found hardly any words were recognised correctly and it would be easier to type in from scratch.

The only free OCR program that I have found, which works well, is Tesseract.

OCR Using Tesseract

(installing Tesseract in SUSE)

gimp

aquire->XSane:Epkowa
grey
dpi=400
gamma=0.67
brightness= -72 (-30 removes reverse side)
contast=100

window:preview
set border

scan

image->mode->indexed
dithering: off
use black&white (1 bit)

save image tiff
compression 1

tesseract in1.tiff out1


metadata block
see also:

 

Correspondence about this page

This site may have errors. Don't use for critical systems.

Copyright (c) 1998-2023 Martin John Baker - All rights reserved - privacy policy.