Malayalam Character Recognition System for Camera Enabled Mobile Devices

Main Article Content

Jomy John
Kannan Balakrishnan, Pramod K. V.

Abstract

In this paper, we propose a character recognition system for mobile camera phones. The system has modules, namely, image pre-processing, image segmentation, feature extraction, and classification. Initially, contrast stretching is performed on the character image captured through mobile camera and it is segmented and stored in the database as a gray scale image. Then, features using different wavelet filters are extracted from gray scale image of characters. For classification, discriminative power of support vector machine is used. The results shows SVM with RBF kernel yield the best performance with 93.12% of accuracy in recognizing isolated characters.


Keywords: character recognition, camera-based image, wavelet features, SVM, RBF

Downloads

Download data is not yet available.

Article Details

Section
Articles