Technique based feature extraction character recognition using artificial neural network for Sinhala characters

Abstract

This paper describes a basic approach, taken for Sinhala handwritten character recognition. The research was performed with the idea of identifying most efficient, effective and accurate method, based on character geometry based feature extraction technique for Sinhala handwritten character recognition. Data acquisition, digitalization, preprocessing, feature extraction was done using the image processing techniques. The classification was measured using an ANN based classifier on a common testing and training data sets. The classification performance was measured for 34 Sinhala characters using this research. Finally, recognized Sinhala character will be printed on a text document.