08-01-2014, 01:13 PM
A Comparative Analysis of Feature Extraction Techniques for Handwritten Character Recognition
ABSTRACT:
Image processing and pattern recognition plays a lead role in handwritten character recognition. There are three main steps of handwritten character recognition- Data collection and pre-processing, feature extraction and classification. In this paper, we have presented different feature extraction methods to classify the 26 handwritten capital alphabets written by 25 different writers with their advantage& disadvantage &compression to each other. Analysis of these feature extraction methods with Back propagation neural network classifier has been done. Neural network is the classifier which we are using for classification with most of the feature vector types.
ABSTRACT:
Image processing and pattern recognition plays a lead role in handwritten character recognition. There are three main steps of handwritten character recognition- Data collection and pre-processing, feature extraction and classification. In this paper, we have presented different feature extraction methods to classify the 26 handwritten capital alphabets written by 25 different writers with their advantage& disadvantage &compression to each other. Analysis of these feature extraction methods with Back propagation neural network classifier has been done. Neural network is the classifier which we are using for classification with most of the feature vector types.