Thursday, 25 August 2011

Optical Character Recognition

What is OCR ?

Optical Character Recognition is a process of extracting text from images. You have noticed that whenever you scan text document ,you can't able to edit it with the help of text editor but with the help of OCR you can
do it

This technology is very important not only for editing scanned documents but for the future of automatic cars.To make a automatic driven car , car should have capability to read instructions written on the street.Moreover, this technology is used by search engines to read what is written on images.

Even though, many believes that it is a solved problem, I don't agree with them.You would not agree with them either after using some OCR software because they still not giving perfect result.  

Related PDF :

Learning on the fly: a font-free approach toward multilingual OCR
Document Specific Modelling

Source Codes :

OCRopus : Open Source OCR Project sponsored by google

OCR code in MATLAB

Python :

gImageReader : A graphical GTK frontend to tesseract-ocr

No comments:

Post a Comment