Combinational domain-based encryption using FrWT and hyper-chaotic system for biometric data security

Abstract

ABSTRACTAn efficient hyperchaos-based combinational domain encryption scheme for securing iris images during transmission over unsecured data networks or database storage is presented in this article. From the region of interest (ROI) of an iris image, distinctive features are strategically extracted for template creation. The significant part, i.e., the ROI of an iris image, is encrypted in a transformational domain and the nonsignificant region is encrypted in a spatial domain. The advantage of this encryption scheme is increased key space and high degree of security compared to 1-D chaos-based systems. Apart from security, another advantage is reduced computational time compared to conventional encryption schemes implemented in either the spatial or frequency domain. The proposed scheme inherits the advantages of increased key space due to transform orders of the fractional wavelet transform (FrWT) and complex dynamical characteristics of hyperchaos, making the proposed scheme resistant to attacks. The properties of the FrWT and hyperchaos complement each other, making the system efficient and highly robust. Experimental results indicate that the proposed scheme reduces the computational time without compromising the security. Further security analysis is done for various attacks such as differential, statistical, and perceptual attacks, which give promising results and indicate that the proposed encryption scheme is efficient and robust.