An Evaluation of Different Feature Extractors and Classifiers for Offline Handwritten Devnagari Character Recognition

Research on Optical Character Recognition (OCR) of Devnagari script is very challenging due to the complex structural properties of the script that are not observed in most other scripts. Devnagari is the script for Hindi, which is the official language of India. Recognition of Devnagari characters poses great challenge due to the large variety of symbols and their proximity in appearance. In this paper, we use two different methods for extracting features from handwritten Devnagari characters, the Curvelet Transform and the Character Geometry, and compare their recognition performances using two different classifiers, viz., the Support Vector Machine (SVM) with Radial Basis Function (RBF), and the k-Nearest Neighbour (k-NN) classifier. Different classification accuracy measures, such as True Positive (TP) Rate, False positive (FP) Rate, Precision, Recall and F-Measure, are used for the purpose. Results obtained show that Curvelet features with k-NN classifier performs the best, yielding accuracy as high as 93.8%.