Handwritten Character Recognition using Perceptual Fuzzy-Zoning and Class Modular Neural Networks
V. L. Lajish · 2007
This paper present a novel feature extraction method for offline recognition of segmented handwritten characters based on the fuzzy-zoning and normalized vector distance measures. Experiments are conducted on forty four basic Malayalam handwritten characters. In the recognition experiments are conducted using class modular neural network with the proposed features and this method is found to be promising.