Abstract

In a multilingual country like India, a document may contain text words in more than one language. For a multilingual environment, multilingual Optical Character Recognition (OCR) system is needed to read the multilingual documents. So, it is segmentation different language. The objective of this project is to propose visual clues based procedure to segmentation Telugu, Hindi and English text portions by line wise, word wise, character wise of the Indian multilingual document. Segmentation is an important topic in script identification and image processing. Previously to identification script by line wise and word wise segmentation poses now am implementation character wise segmentation it is easy way to identification language by character3 wise segmentation. The objective of segmentation is to segmentation by image line wise, word wise, charter wise. The segmentation is very important role to identification of languages. The world we live in, is getting increasingly interconnected, electronic libraries have become more pervasive and at the same time increasingly automated including the task of presenting a text any language as automatically translated text in any other language. Identification of the language in a document image is of primary importance for selection of a specific OCR system processing multi lingual documents.

Authors and Affiliations