[Anubad] Parichit ( परिचित ) is an attempt to create best open source OCRs for Indian and South Asian languages.

[sankarshan mukhopadhyay]

<http://code.google.com/p/parichit/>

Parichit ( परिचित ) is an attempt to create best open source OCRs for Indian and South Asian languages. The project is based on Tesseract OCR Engine ( http://code.google.com/p/tesseract-ocr/ ) . The front end will be a modified version of VietOCR ( http://vietocr.sourceforge.net/ )

The project aims to create high quality training data for creating tesseract language models for each of the Indian languages.

Shirorekha
Devanagari, Bengali and Gurmukhi scripts are written in a way where each word is marked by a line above it called as Shirorekha. This connects all the letters in a word and so segmetation had to be done before the scanned is passed on to the recognizer.

A Shirorekha segmenter will be made available on this project soon.

Character Reordering
Almost all Indic scripts need character reordering to reorganise from visual order to logical (Unicode) order. Since Tesseract OCR operates strictly from left to right the characters are scanned in visual order and recognition also happens in visual order. This needs to be reordered in post processing. Code for such reordering for each language will be made available soon.

-- 
sankarshan mukhopadhyay
<http://sankarshan.randomink.org/blog/>