Novel image encryption algorithm using fractional chaos and cellular neural network

The work presented in this article gives a novel digital image encryption algorithm using a fractional-order chaotic system and cellular neural network. The encryption is done on the lines of chaos-based permutation-substitution architecture. The main contribution lies in key-generation, which is inspired from the Merkel-Damgard scheme. The diffusion mechanism is performed with the help of Conway's game of life and NARX network. Extensive experimental results of the cipher indicate that the algorithm can withstand classical cryptanalysis and can outperform many other existing image encryption algorithms. Moreover, the scheme fairs pretty well on theoretical aspects of security and passes most of the NIST standards.