Intelligent Algorithm In Biometric Identification Technology

The recent advances of information technology and theincreasing requirement for security have resulted in a rapiddevelopment of intelligent personal identification based onbiometrics. Biometric Identification Technology is a kind ofsciences of using individual personal characteristics toverify identity. It represents the most secure way to identifyindividuals because it provides a novel approach to recognizethe identity, or verify the claimed identity through aunique, highly reliable and robust physical characteristic orpersonal trait. Generally, biometrics include the following:facial feature, iris, gait, voiceprint, gesture, fingerprints,handwritten signature and so on. On the other hand, This paperproposed a novel cooperative evolutionary algorithm based onsimulated annealing algorithm(SA) and particle swarmoptimization(PSO).This new method utilizes the globalconvergence property of SA and the facility of realization ofPSO, while it attains an optimal solution by means of theinformation exchanges of these two kinds of algorithms. Resultsof experiments show that the proposed algorithm performedbetter in convergence speed and precision of solutions than GA, PSO, and SA. It is an efficient method in BiometricIdentification Technology.This article is divided into four parts. The first chapterintroduces the background and application of the BiometricIdentification Technology and the latest research in the world.The second chapter briefly introduces the research historyof Simulated annealing algorithm and Particle Swarm Optimization algorithm and biological background at the beginning.Then proposed a novel cooperative evolutionary algorithm basedon simulated annealing algorithm(SA)and particle swarmoptimization(PSO).This new method utilizes the globalconvergence property of SA and the facility of realization ofPSO, while it attains an optimal solution by means of theinformation exchanges of these two kinds of algorithms. Resultsof experiments show that the proposed algorithm performedbetter in convergence speed and precision of solutions thanGA, PSO, and SA.In third chapter, we apply the former algorithm inBiometric Identification TechnologyThe forth chapter gives a lots of examples, which prove thatthe algorithm we present is feasible and simple, It is anefficient method in Biometric Identification Technology.