Title: HANDWRITTEN DEVANAGARI (MARATHI) CHARACTER RECOGNITION USING SVM AND MLP
Abstract: Many researchers are working to automate the process of reading, understanding and interpretation of handwritten character. In this paper, we present the new machine learning algorithm for Handwritten Devanagari (Marathi) characters recognition. Offline handwritten character recognition is an important area of Document Analysis and Recognition (DAR). DAR is a mechanism in which the document images are processed to obtain features and its recognition. Each character is stored as an image. The histogram oriented gradient features are extracted. The algorithm is designed to select the extracted features. There is 25% reduction in the feature set. These features are classified using SVM and the obtained average recognition accuracy is on testing is 95.83% and on cross validation is 95.82%. MLP classifiers accuracy is 95.45% on testing and on cross validation is 95.32 % with TanhAxon as Transfer function.
Publication Year: 2015
Publication Date: 2015-01-01
Language: en
Type: article
Access and Citation
Cited By Count: 2
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot