Recognition of Offline Handwritten Odia Numerals Using Support Vector Machine
Pushpalata Pujari, Babita Majhi · Computational intelligence · 2015
Extensive work has been done on recognition of languages like Hindi, Kannada, Tamil, Bangala, Malayalam, Gurumukhi, Arabic, Farsi and Chinese etc. But a very few work has been done on Odia character Recognition. There are many fields like banking, postal system, and form processing etc. Which require effective digit recognition system for faster processing. In this paper a sincere attempt has been made to do a comparative study using different types of classifiers for handwritten Odia numerals. For feature extraction gradient and curvature based approaches are used. After the calculation of feature vector Principal Component Analysis (PCA) is applied to reduce the size of feature vector. The reduced features are passed to SVM classifiers. The result obtained is compared with other classifiers such as Multilayer Neural Network (MNN), Decision tree(C5.0)and Discriminant Analysis (DA). From the experimental result it is observed that SVM based classifier achieved 90.5% accuracy with Curvature feature and 95.5% accuracy with gradient feature during validation.