Chaotic image encryption based on spiral traversal and finite field bidirectional diffusion

Abstract

In this paper, a new sinusoidal cascade chaotic map model is proposed, which constructs chaotic system by sinusoidal cascade of two identical 1D chaotic maps. This model can generate 1D chaotic system with wider parameter range and wider chaotic region. To verify its application in image encryption, a chaotic image encryption algorithm based on spiral traversal and finite field bidirectional diffusion is proposed. The pseudo-random sequences generated by three kinds of 1D chaotic maps improved by the sinusoidal cascade chaotic map model are used as the ciphers of the cryptosystem. Among them, SHA-256, SHA-512 and plain image are combined to generate the initial value of the system. The experimental results and security analysis show that the encryption algorithm designed in this paper can effectively resist exhaustive attack, differential attack, noise attack and so on, and the proposed model has superior performance in encryption speed and security.