Optical Character Recognition

"tesseract-ocr" is a good free program for optical character recognition. I used it for generating German text from screenshots. Here is how to proceed:

Installation

• Learn about tesseract-ocr: tesseract/wiki
• Download it from github.com/UB-Mannheim
  I got the version 3.05.
• Download the language file "deu.traineddata" for German from tessdata/tree.

Usage

• Use cygwin on Windows
• Get help:
  $PATH/tesseract.exe -h
  $PATH is the path to the directory with tesseract.exe
• Generate text from image:
  $PATH/tesseract.exe test1.tif test1 -l deu

Kommentare