Author(s): Tanmay Bhattacharya | Sirshendu Hore | S. R. Bhadra Chaudhuri
Journal: International Journal of Computer Science Issues
ISSN 1694-0784
Volume: 8;
Issue: 5;
Start page: 209;
Date: 2011;
Original page
Keywords: ANN; Minutiae; Center of Gravity;Aspect Ratio; Training; SHA-512 | Crossover | IJCSI
ABSTRACT
Unimodal biometric systems have to contend with a variety of problems such as noisy data, intra-class variations, restricted degrees of freedom, non-universality; phishing attacks spoof attacks, and high false acceptance rates. In order for the biometrics to be ultra-secure and to provide more-than-average accuracy, more then one form of biometric identification is required. Hence some of these limitations can be addressed by a combination of different biometric recognition technologies that integrate the evidence presented by multiple sources of information. In the proposed work Fingerprint Key and Signature Keys are generated from the Fingerprint and Handwritten Signature of the legitimate user. The system is quite robust because it is trained by Artificial Neural Network and Machine Intelligence. Two Combined Keys are generated by Genetic crossover of those two keys. Finally by interleaving the combined keys Encryption Key is generated. In this approach there is a significant improvement over the traditional Unimodal biometric authentication techniques.
Journal: International Journal of Computer Science Issues
ISSN 1694-0784
Volume: 8;
Issue: 5;
Start page: 209;
Date: 2011;
Original page
Keywords: ANN; Minutiae; Center of Gravity;Aspect Ratio; Training; SHA-512 | Crossover | IJCSI
ABSTRACT
Unimodal biometric systems have to contend with a variety of problems such as noisy data, intra-class variations, restricted degrees of freedom, non-universality; phishing attacks spoof attacks, and high false acceptance rates. In order for the biometrics to be ultra-secure and to provide more-than-average accuracy, more then one form of biometric identification is required. Hence some of these limitations can be addressed by a combination of different biometric recognition technologies that integrate the evidence presented by multiple sources of information. In the proposed work Fingerprint Key and Signature Keys are generated from the Fingerprint and Handwritten Signature of the legitimate user. The system is quite robust because it is trained by Artificial Neural Network and Machine Intelligence. Two Combined Keys are generated by Genetic crossover of those two keys. Finally by interleaving the combined keys Encryption Key is generated. In this approach there is a significant improvement over the traditional Unimodal biometric authentication techniques.
