In this blog, i would share FOSS related things that i come across in my daily life.

Wednesday 25 June, 2008

Free / Open Source OCR software

I came across a nice, easy and free OCR software.

The name is "Softi Free OCR Software"
http://www.softi.co.uk/freeocr.htm




















I have personally found it very accurate and free.
It had a downside in version 1.0, as you were not allowed to perform OCR on a portion of image.
In version 2.4, that problem is not more as you can easily select a portion of image on which OCR is to be performed.

I wish it had automatic layout detection. At least the ability to identify that the image contains two page scans would be really useful when it comes to scanning from paper book.

I hope in future release, they would add the functionality of performing OCR on images in PDF document. That would be really handy.

P.S. :: The software itself may not be open source, but it doesnt matter much as it is only a GUI frontend to popular command-line OCR engine name Tesseract.
http://code.google.com/p/tesseract-ocr/