Image encryption using asynchronous non-uniform life-like cellular automata and a fractional-order generalized 3D Hénon map

This paper presents a chaos-assisted reversible encryption scheme for RGB images based on asynchronous non-uniform Life-like cellular automata and second-order bit-plane diffusion. The input color image is first decomposed into its three channels and processed independently in a block-wise manner. A key-dependent fractional-order generalized 3D Hénon chaotic generator is used to control local pixel disturbance, block permutation, bit-plane shuffling, local rule assignment, asynchronous update order, and diffusion iterations. To enhance diffusion while preserving exact invertibility, a reversible second-order formulation is embedded into the bit-plane evolution process. Moreover, a balanced, complementary Life-like rule pair is employed to promote more uniform binary-state transition behavior. Each channel undergoes chaos-driven local pixel disturbance, block permutation, reversible asynchronous non-uniform cellular automata diffusion, and forward–backward inter-block chaining before the encrypted channels are recombined to produce the final cipher image. Experimental and analytical results demonstrate that the proposed scheme generates noise-like cipher images with an entropy of 7.9996, close to the ideal value of 8. The adjacent-pixel correlation coefficients are 0.0002, -0.0016, and 0.0016 in the horizontal, vertical, and diagonal directions, respectively. In addition, the proposed method achieves an NPCR of 99.6100% and a UACI of 33.4521%, indicating strong diffusion capability and resistance to differential attacks. The proposed method supports exact lossless decryption and shows strong resistance to statistical, differential, ciphertext-only, known-plaintext, and chosen-plaintext attacks, confirming the effectiveness and robustness of the proposed framework for secure RGB image encryption.