Performance Comparison of SVM and ANN for Handwritten Devnagari Character Recognition

TL;DR

This paper compares the performance of Support Vector Machines and Artificial Neural Networks in recognizing handwritten Devnagari characters, highlighting feature extraction and decision fusion techniques for improved accuracy.

Contribution

It presents a comparative analysis of SVM and ANN methods for Devnagari character recognition, including feature extraction and decision fusion strategies.

Findings

SVM and ANN achieve competitive recognition accuracies

Feature extraction methods significantly impact classification performance

Decision fusion improves overall recognition accuracy

Abstract

Classification methods based on learning from examples have been widely applied to character recognition from the 1990s and have brought forth significant improvements of recognition accuracies. This class of methods includes statistical methods, artificial neural networks, support vector machines (SVM), multiple classifier combination, etc. In this paper, we discuss the characteristics of the some classification methods that have been successfully applied to handwritten Devnagari character recognition and results of SVM and ANNs classification method, applied on Handwritten Devnagari characters. After preprocessing the character image, we extracted shadow features, chain code histogram features, view based features and longest run features. These features are then fed to Neural classifier and in support vector machine for classification. In neural classifier, we explored three ways of combining decisions of four MLP's designed for four different features.