Tamil Character Recognition, Translation and Transliteration System

Abstract

Optical Character Recognition (OCR) is the machine conversion of handwritten or typed data into machine encoded scripts. English, is the informal link to all the regional languages in India and is used to publish reports, papers, magazines and records. Monolingual systems are incapable of doing so, thus increasing the need of bilingual frameworks. We have created a character recognition system that converts the user’s input in the Tamil language to English. Additionally, we can also perform transliteration of Tamil to English and vice versa. The bilingual OCR system, “MOZHI VALLAAN,” which has an accuracy of 94%, will be used.