Visual double image encryption scheme based on multilayer complex networks and wavelet transform

This paper explores the synchronizability within a three-layer network, which has the feature of identical topologies across layers but with three different kinds of inner coupling patterns, namely, injective, radiative, hybrid. A fractional-order chaos is designed and employed as the nodes of the network, which is then applied to image encryption combining with hash function and improved Zigzag transform. It is shown that the three-layer network can achieve inter-layer synchronization under specific conditions. Based on the fractional-order three-layer complex network and the improved Zigzag transformation, a robust symmetric encryption algorithm is designed, leveraged by the classical scramble-diffusion mechanism. After employing the fractional-orders and increasing the network nodes of the complex network, the key security is significantly increased. Theoretical analyses and experimental results show that the scheme has bigger key space and stronger key sensitivity, which can resist typical attacks effectively. Furthermore, the secret image is concealed within the host image and transmitted to the receiver via internal wavelet transformation. Through comparative analyses of various algorithms, the visually double encryption scheme exhibits superior encryption efficacy for digital images.