Comparison of Offline Handwritten Character Recognition Using Wavelet And Contour let Feature Extraction Technique

Abstract

Feature extraction in an important process in character recognition, multiresolution techniques play important role in extracting the feature from the input image. In this paper multiresolution techniques such as wavelet and contourlet is used for comparison. Image processing with artificial neural network is used to recognition the offline handwritten characters. In offline character recognition, processing of various steps in series are processed, such as pre-processing, dilation feature extraction & finally classification is done using artificial neural network. In preprocessing we increase the quality of the image by reducing the noise. The feature estimated using wavelet and contourlet is saved for the each image in the training set. Thus the saved features are given as input to the feed forward Back Propagation Artificial Neural Network, that produces the output based on the trained images.