CNN Based Recognition of District Names of Punjab State in Gurmukhi Script

Loading...
Thumbnail Image

Date

item.page.authors

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

From the last few years, numbers of Researchers all over the world have been working for the automation of Postal system. Development of Postal Automation for a Country like India is difficult as compared to other countries. As India has multi-lingual and multi-script behavior which makes the development of Postal Automation System a more difficult. Sometimes, the postal document to be posted is written in three languages like English, Hindi and in the state s official language. The work carried out in this thesis addresses the problem of handwritten word recognition for the development of postal automation for the state Punjab. Postal Automation is already available for the various scripts including English, Bangla, Malayalam and Devanagri etc. For the development of Postal automation for the state Punjab, all the 22 district names which are written in Gurmukhi script is taken into consideration. Gurmukhi script is the official language of the state Punjab and being used by all the government officials to post their documents. The approach used for the purpose of Gurmukhi district names recogition is based on the Holistic approach instead of Analytical approach by employing Deep learning algorithm(CNN), in which features are automatically extracted from the complete word without segmentation. This makes the used approach better as compared to previously used approaches. This thesis is divided into five chapters. A brief outline of each chapter is given in the following paragraphs. newline

Description

Keywords

Citation

item.page.endorsement

item.page.review

item.page.supplemented

item.page.referenced