In a multilingual country like India, a document may contain text words in more than one language. It is reasonably natural that the documents produced at the border regions of Karnataka may also be printed in the regional languages of the neighbouring states like Telugu, Tamil, Malayalam and Urdu. An electronic Searching system for Kannada text, based on the content is needed to access such multilingual documents. So, it is necessary to identify different language portions of the document before the retrieval. The objective of this paper is to propose visual clues based procedure to identify Kannada text from a multilingual document w hich contains Hindi, English and Malayalam text portions along with Kannada
展开▼