Use of Dominant Point Detection Feature for Recognition of Online Handwritten Devanagari Script

Abstract

After the invention of computers, a great amount of work has been done in the field of computer human interface. But the problem of exchanging data between human beings and computing machines is still challenging. An enormous research has been done for efficient character recognition and even now research is going on. Basically Character recognition techniques associate a symbolic identity with the image of a character. To make the complicated process of online handwritten character recognition easier and more robust, focus should be on salient features of character. After pre-processing feature extraction is done. Feature extraction is a vital phase of character recognition. Features extracted from character encode the structural characteristics of character shape. One of the methods to get efficient recognition is to extract dominant points of characters. Dominant points can be used to recognize the character more proficiently. Dominant points are commonly considered as points with local maximum curvature (elevated position). In other words, it can be said that dominant points of a character are those points where the slope value changes noticeably. The dominant points are taken as output from feature extraction phase and input for recognizing the character. The dominant points are extracted for characters and also the distance of these points from centre of character is calculated. On the basis of these features, recognition is accomplished. Recognition is done through SVM.