Enhancing AES image encryption with a three-dimensional hyperchaotic system for increased security and efficiency

In the digital era, the security of images, as critical carriers of information, is paramount for national security, military strategy, and personal privacy protection. Therefore, developing efficient and cost-effective image encryption technologies to ensure the security of image data during transmission has become an urgent necessity. Although the Advanced Encryption Standard (AES), a widely used symmetric encryption method, performs excellently in data communication and network security, its efficiency and security face significant challenges when directly applied to image encryption due to the inherent complexity of image data. This paper presents a simplified AES image encryption framework based on a three-dimensional hyperchaotic system (TDHCS). The crux of this framework is the incorporation of a novel TDHCS, distinguished by its intricate nonlinear dynamics and robust randomness. The AES encryption process is simplified based on the high-level random chaotic sequences generated by TDHCS. In contrast to the traditional fixed S-box, a dynamic S-box generation scheme is employed, while a random operation scheme is proposed as a substitute for fixed sequences in the round key addition and row shifting steps. Experimental results show that the proposed algorithm reduces encryption time to 2.2369 s for a 256×256 grayscale image, representing an 87.08% improvement over traditional AES (17.3090 s) and a 48.41% improvement over other AES-based chaotic encryption algorithms (4.3360 s). Furthermore, the security analysis demonstrates that the algorithm effectively resists differential attacks, achieving NPCR, UACI, and BACI values with deviations of only 0.0031%, 0.0046%, and 0.0288%, respectively. The proposed algorithm markedly reduces the number of encryption rounds to one while enhancing both efficiency and security. Simulation results confirm its robustness against various cryptographic attacks, demonstrating its potential as a preferred solution for digital image privacy protection.