Research On Multimodal Handmetric Recognition

Biometrics is the most secure and convenient way to satisfy the requirements for identity digitalization and virtualization in the coming network society, which refers to the automatic identification of an individual by using certain physiological or behavioral traits associated with person. Multimodal biometrics system is more robust than unimodal one. It takes advantages of multiple biometric traits to improve the performance in many aspects including accuracy, noise resistance, universality, spoof attacks, and reduce performance degradation in huge database applications. Recently, new algorithms and applications of multimodal biometrics are emerging rapidly.Multimodal biometrics technology is a new direction of the biometrics technology, in which there are a lot of problems need to be studied and solved. All of multimodal biometrics technologies are facing two major issues: how to pick up the optimal biometrics combination which can meet actual demands on engineering application and what kind of strategy should be chosen for the fusion of Multimodal biometrics on the academic study. This paper is focusing on how to solve these two problems,which is funded by Chongqing Municipal Education Commission.Hands are selected to be the subjects of multimodal biometric in this paper, which have abundant biometric characteristics. Firstly, fundamental researches of all kinds of handmetric technologies are given in the paper. On this basis, a simple and practical mathematical model is built up in this paper, which utilizes both multi-goal programming and 0-1 linear programming techniques in optimization theory. The specific MATLAB program corresponding to the model is also given in the paper. The simulation experiment indicates that an optimal biometric features combination which satisfied a bundle of restrictive rules can be given out rapidly and effectively by the optimization model proposed in this paper.A specific biometric features combination ("finger vein and finger creases") is selected by this optimization model, which is taken as the object of study on fusion strategies of multimodal handmetrics from three levels (data layer,feature layer and decision layer) of information fusion. Novel multimodal fusion strategies on three levels of fusion and corresponding experiments are given in this paper. The theoretical researches and experiments indicate that the multimodal biometrics can get little benefit in the fusion on data layer, but has more research value and better fusion effects on feature layer and decision layer.