Conception of Biometric System Based on Optimization of a Fuzzy System

Abstract

Recently, user identification is an essential foundation for protecting information in several applications. To overcome the classical security means, a natural and reliable solution to this problem is offered by biometric technologies. So among several biometric modalities, these extracted from the hand, e.g. palmprint have been systematically used to make identification for the last years. In this work we proposed a new feature extraction algorithm for biometric identification purpose. The characteristic vector is a collection of fuzzy data base rule parameters obtained through optimization of error target function. For that Newton Raphson method is used to establish the proposed algorithm. Thus, our contribution is the introduction and the development of adaptive fuzzy system as biometric image modeler. Then for the evaluation of this method we presented Poly U multispectral palmprint database and we exanimated the proposed method in both cases before and after the data fusion. The results showed that our method is fast with high identification accuracy and can be used in high secure application.