An investigation of ensemble systems applied to encrypted and cancellable biometric data

Abstract

In this paper, we propose the simultaneous use of cryptography and transformation functions in biometric-based identification systems aiming to increase the security level of biometric data as well as the performance of these systems. Additionally, we aim to keep a reasonable efficiency level of these data through the use of more elaborated classification structures, such as ensemble systems. With this proposal, we intend to have a robust and secure identification system using signature data. © 2012 Springer-Verlag.