OCR software
Google has an open source OCR software called: tesseract-ocr http://code.google.com/p/tesseract-ocr/
Seems there’s a problem with the code, at least in Visual Studio, it does not build successfully. Not sure if it’s true for its Linux version(Tried recently, it works!).
Instead, there’s a free(?) software, which is basically a wrapper for this, called Free OCR: http://softi.co.uk/freeocr.htm