Handwritten Devanagari Lipi using Support Vector Machine
Shailendra KumarShrivastava, Pratibha Chaurasia · International Journal of Computer Applications · 2012
The handwritten recognition is a one of the basic biometric recognition technique.Different technique and features are used for the faithful recognition characters.In this paper we have proposed a SVM (support vector machine) based technique for Devanagari character recognition.The Devanagari characters have very correlation to each other.This feature of the Devanagari lipi make difficult to faithful recognition.The energy features of segment characters are used for the classification.The more no. of segmentation improves the recognition rate.The different recognition rates with no. of segment are used in this paper.The recognition rate is also developed on the kernel of SVM.The result of different kernel is also given in this paper.