The past few years has concentrated research on OCR for Chinese, Japanese and many language scripts. A lot of work has been also reported on OCR for various Indian scripts, which are Devanagari, Malayalam, Bangla, Kannada, Gurumukhi, Gujarati, Oriya, Tamil, and Telugu etc. In this paper, we present our work using OCR on Indian scripts, mainly on Devanagari, the most popular script in India.
Our work deals with the recognition of only the offline handwritten compound characters. In this work handwritten compound character recognition in scanned documents is introduced. There are many research paper presented a new perspective which improve the current system along with experimental results. The proper selection of the technique is most probable for achieving high performance with a better accuracy.