A novel multi-chaos and DNA-based security system for multi-image encryption across diverse applications

Abstract In today’s high-volume communication environment, large numbers of images are transmitted over open networks, making the secure and efficient protection of multiple images an important challenge. Conventional single-image encryption (SIE) techniques often become inefficient or less robust when directly extended to multi-image scenarios and typically do not integrate pixel-level confusion and bit-level diffusion in a unified framework. To address these limitations, this paper proposes a multi-image encryption system based on DNA encoding, multiple one-dimensional chaotic maps, and SHA-256-based key generation. The hash of the combined plaintext images is used to derive image-dependent initial conditions and parameters, ensuring strong key sensitivity. Multiple chaotic maps are employed to generate permutation and rule-selection sequences, while dynamic DNA encoding and two-stage chaotic shuffling provide enhanced confusion. A DNA-XOR diffusion stage further strengthens bit-level security. The combined design increases unpredictability and resistance to statistical, chosen-plaintext, and chosen-ciphertext attacks. Experimental results show that the proposed method achieves high security performance, with information entropy values close to 8, NPCR above 99.62%, UACI around 33.47%, near-zero adjacent-pixel correlation in cipher images, and competitive execution time. These results demonstrate that the proposed scheme provides secure and efficient multiple-image encryption suitable for practical transmission applications.