Tamil Handwritten Character Recognition Using Back Propagation Network

The extracted features for recognition are converted to back propagation network (BPN) where the characters are classified using Fourier descriptors algorithm. In this paper we propose an approach to recognize handwritten Tamil characters using a multilayer perceptron with one hidden layer. The feature extracted from the handwritten character is Fourier Descriptors. Also an analysis was carried out to determine the number of hidden layer nodes to achieve high performance of back propagation network in the recognition of handwritten Tamil characters. The system was trained using several different forms of handwriting provided by both male and female participants of different age groups. Test results indicate that Fourier Descriptors combined with back propagation network provide good recognition accuracy of handwritten Tamil characters.