Handwritten Character Recognition using Morphological Operators with Competitive Neural Trees

Abstract

In this paper an attempt is made to develop Myanmar handwritten character recognition system. Character recognition is an important area in image processing and pattern recognition fields. The aim of character recognition is to translate human readable characters to machine readable characters. The paper describes the process of character recognition using morphological operators with the competitive neural trees. The morphological operators are used to extract the edge of each character image. Then, competitive neural trees (CNeT) are used for classification. It is one of the fast supervised neural networks with high performance. The main advantage of the CNeT is its structured, self-organizing architecture that allows for short learning and recall times. High speed recognition rates can be gained by using CNeT.