A novel signature verification and authentication system using image transformations and Artificial Neural Network

This paper proposes an Artificial Neural Network based approach for implementing Automatic Signature verification and authentication system. In this era, with the rapid growth of Internet and the necessity of localized verification systems, handwritten signature has become an important biometric feature for the purpose of verification and authentication. The proposed method comprises spatial and frequency domain techniques for transformation. After extracting the Region of Interest Ripplet-II Transformation, Fractal Dimension and Log Polar Transformation are carried out to extract descriptors of the concerned signature to be verified as well as authenticated. In decision making stage Feed Forward Back Propagation Neural Network is used for verification and authentication purpose. This system has been tested with large sample of signatures to show its verification accuracy and the results have been found around 96.15%. Also forgery detection rate has been found 92% which is very encouraging. False Acceptance Rate and False Rejection rate of our system has been determined 5.28% and 2.56% respectively. This approach has been compared with some existing system and it has been observed that this system shows better performance.