Chaos based fingerprint image encryption in the wavelet domain

Abstract

With the widespread of distributed biometric-based authentication systems, security of the biometric data transmitted over open networks has become a major concern, as there are higher risks of identity theft and replay attacks. In this paper, a new chaos-based encryption scheme that operates in the Integer Wavelet Transform domain is proposed to enhance the security of fingerprint images during transmission. The proposed scheme relies on combining the chaotic Lorenz system, blocks permutation and the Rubik’s cube principle to achieve confusion and diffusion operations. Experimental results show that the proposed technique provides high performances in terms of security, computation time and quality of encrypted images.