A Neural Network Approach to Character Recognition

In this paper, an attempt is made to develop off-line recognition strategies for the isolated Handwritten English character (A to Z) and (0 to 9). Challenges in handwritten character recognition wholly lie in the variation and distortion of handwritten characters, since different people may use different style of handwritten, and direction to draw the same shape of the characters of their known script. The paper provides a review on the process of character recognition using neural network. Character recognition methods are listed under two main headlines. The Offline methods use the static images properties. The Offline methods are further divided into four methods, which are clustering, Feature Extraction, Pattern Matching and Artificial Neural Network. The Online methods are subdivided into k-NN classifier and direction based algorithm. Character preprocessing is used binarization, thresolding and segmentation method. Neural network based method improves the character recognition. The proposed method is based on the feed forward back propogation method to classify the characters. The ANN is trained using the Back Propogation algorithm. In the proposed system, English nue-merical letter is represented by binary numbers that are assume as input and fed to an ANN. Neural network followed by Back Propagation Algorithm which compromises Training.