Recognition of Alphanumeric Patterns Using Backpropagation Algorithm for Design and Implementation With ANN

Abstract

The artificial neural network has been called for its application as alphanumeric characters recognizing the network. The idea is to maintain the obsolete data available in hard copy form and to convert that data into digital form. Some specific bit patterns that correspond to the character are trained using the network. The numbers of input and output layer neurons are chosen. As there are many ways to do but in this the algorithm used for training the network is called the Backpropagation Algorithm using the delta rule. The testing and training patterns are provided for which weights are calculated in the program and patterns are recognized and analysis is done. The effect of variations in the hidden layers is also observed with pattern matrices.