Typed Malayalam Character Recognition

Main Article Content

Amalu Johns
Jibrael Jos

Abstract

Malayalam is an Indian language .It is mainly using in Kerala one of the states in India. It is very difficult that to recognize Malayalam characters because it contains large number of curls, curves, loops etc... Many characters seem similar to each other. Many OCR systems are already available which can convert the printed or handwritten characters into machine encoded form but comparatively less number of systems are there to identify Malayalam characters. Here the recognition is performed through many stages. Starting from image acquisition for taking the input images, then the main task is that to remove noise from the images and thinning them. Thinning is performed to easily recognize each pixel from the character. Segmentation and feature extraction are the main procedures used for this. Based on the extracted features, characters are classified. The main application of this procedure is that it will help in effortlessly understanding the pronunciation of a word in different languages. This system obtains typed Malayalam characters from the internet and converts them into images which are then analyzed to identify the character.


Keywords: pre-processing, feature extraction, classification, segmentation

Downloads

Download data is not yet available.

Article Details

Section
Articles