Chaotic image encryption algorithm based on block theory and reversible mixed cellular automata

Abstract

In this paper, a new chaotic image encryption algorithm based on block theory and reversible mixed cellular automata model is proposed. Different from traditional confusion-diffusion structure, this paper employs new three-tier architecture, i.e. diffusion-confusion-diffusion. The first diffusion operation adopts the traditional diffusion method, which aims to avoid direct operation on plaintext and further enhance the encryption effect. The next scrambling and diffusion operations are based on image blocks and a new dynamical system, namely reversible mixed cellular automata model. The scrambling operation is divided into two parts, inter block scrambling and intra block scrambling, all the above operations are based on pixel level. Then, diffusion operation is performed and its implementation is based on the evolution of constructed reversible mixed cellular automata with balanced rules. In addition, the initial values and parameters of the adopted chaotic system, two dimensional (2D) Logistic map, are obtained using SHA 256 hash function related to plaintext image. The use of hash function makes the chaotic sequences completely different when encrypt different images, which enhances the ability to resist known-plaintext and chosen-plaintext attacks. Simulation results and performance analysis bear out that the encryption algorithm proposed in this paper has higher security performance and can resist different attacks efficiently.