Classifier Fusion Method to Recognize Handwritten Kannada Numerals
Optical Character Recognition (OCR) is one of the important fields in image processing and pattern recognition domain. Handwritten character recognition has always been a challenging task. Only a little work can be traced towards the recognition of handwritten characters for the south Indian languages. Kannada is one such south Indian language which is also one of the official language of India. Accurate recognition of Kannada characters is a challenging task because of the high degree of similarity between the characters. Hence, good quality features are to be extracted and better classifiers are needed to improve the accuracy of the OCR for Kannada characters. This paper explores the effectiveness of feature extraction method like run length count (RLC) and directional chain code (DCC) for the recognition of handwritten Kannada numerals. In this paper, a classifier fusion method is implemented to improve the recognition rate. For the classifier fusion, we have considered K-nearest neighbour (KNN) and Linear classifier (LC). The novelty of this method is to achieve better accuracy with few features using classifier fusion approach. Proposed method achieves an average recognition rate of 96%.