Private chaotic biometric template protection algorithm

Biometric authentication systems have proven a priority over traditional systems like passwords. But while biometrics ensure uniqueness, they do not provide secrecy and security of the users data. Many attacks can be launched against biometric systems resulting in the reduction of these systems credibility. Specially, the attack on the biometric templates stored in the system database consists an attack on the integrity of these templates. In this paper we present the design and development of a new approach for biometric template protection based on the chaotic behavior of logistic map. The proposed biometric template protection algorithm satisfies the requirements of revocability, diversity and privacy. Moreover, experimental results conducted on several face databases show the ability of the proposed scheme to preserve the performance of the protected system.