OFF-LINE HANDWRITTEN SIGNATURE VERIFICATION USING AN OPTIMUM CHARACTERIZATION METHOD

Automatic signature verification is a well-established and an active area for research with numerous applications such as bank check verification, ATM access, etc. Most off-Line signature verification systems depend on pixelsf intensity in feature extraction process which is sensitive to noise and any scale or rotation process on signature image. This paper proposes an off-line handwritten signature verification system using a combination model among Discrete Wavelet Transform, Wavelet Neural Network and Genetic Algorithms for optimizing feature extraction and verification methods. Discrete Wavelet Transform is used to extract wavelet energy and wavelet entropy values from each signature image without any dependency of pixels intensity. Since each signature should be characterized by an optimal wavelet analysis function, Genetic Algorithms, searching strategy; select the most effective wavelet analysis function and entropy parameters to each signature. Because Discrete Wavelet Transform suffers from down-sample process, Wavelet Neural Network is used as a classifier to solve this problem. Since each signature should be characterized by an optimal wavelet activation function, Genetic Algorithms is used to select the most effective wavelet activation function and learning rate value to each signature. The details and limitation of this paper will be discussed where a comparative study should be appear. Moreover, conclusions and future works would be discussed