A multi-layer windmill image retention encryption algorithm based on composite sequences and DNA
Ning Zhang, Mingxin Fang, Bing Zhao · Journal of King Saud University - Computer and Information Sciences · 2026
With the widespread application of multimedia information technology in cloud storage and secure communications, the secure transmission and storage of digital images have become a critical challenge.To address security challenges in digital images, this paper proposes a multi-layer windmill image encryption scheme integrating multidimensional chaos, quantum-inspired random walk, and DNA encoding. This approach employs chaotic sequences to drive quantum-inspired random walks for generating composite keys, thereby resolving the pseudo-randomness trap inherent in traditional chaotic encryption. By integrating multi-rule DNA encoding with mathematical matrix operations, it overcomes the limitations of single-base pairing, namely insufficient randomness and linearity defects. Furthermore, a multi-layer windmill-style error correction structure is designed to achieve 3/4 cropping attack resistance while enabling effective image reconstruction. Experiments demonstrate that this scheme possesses a key space exceeding 2 25 ⁶, information entropy approaching 8, and high sensitivity to both plaintext and keys (NPCR≈99.60%, UACI≈33.46%). It effectively resists statistical analysis, differential attacks, noise, and cropping attacks, exhibiting superior security compared to mainstream comparative algorithms.