An Image Encryption Algorithm Based on a New Fractional Order Chaotic Neural Network
Nanming Li, Shu-Cui Xie, Jianzhong Zhang, Yangguang Lou · 2022 4th International Conference on Natural Language Processing (ICNLP) · 2022
An image encryption algorithm based on a new fractional order chaotic neural network (CNN) is proposed. Firstly, a three-dimensional continuous integral order CNN is obtained by numerical simulation on the chaotic neuron model. Then, the integral order CNN is extended to fractional order, and the fractional order CNN iteratively generates pseudo-random sequences for subsequent encryption. In addition, the initial value of the fractional order CNN is generated from the hash value (secure hash algorithm: SHA-256) of the plain image. Finally, we apply the encryption structure of forward diffusion, scrambling and backward diffusion to the encryption algorithm, and diffusion is performed at the bit level, scrambling is performed at the pixel level and bit level. Experimental results and security analysis show that the algorithm has better performance and can resist typical attacks.