Abstract
Securing visual information against sophisticated cyber threats remains a core challenge in modern cryptography because many existing chaos-based image encryption schemes suffer from low key sensitivity and static substitution. To overcome these intrinsic limitations, this study develops a multistage image encryption framework by synergistically fusing the Lorenz chaotic system, Secure Hash Algorithm 256 (SHA-256), and Discrete Time Quantum-inspired Walks (DTQWs). The chaotic Lorenz system yields highly sensitive diffusion sequences via bitwise modular operations, whereas the DTQW dynamically constructs plaintext-dependent Substitution Boxes (S-Boxes) and thereby reinforces confusion to minimize statistical predictability. The SHA-256 hash introduces a session-dependent quantum coin rotation parameter to ensure dynamic evolution with intrinsic plaintext sensitivity during the encryption process. Extensive simulations demonstrate outstanding security performance of the proposed scheme: near ideal entropy value of 7.9999, the Number of Pixels Change Rate (NPCR) and the Unified Average Intensity Value (UACI) rates of [Formula: see text] and [Formula: see text], correlation coefficients close to zero, and high decryption reconstruction fidelity with Peak Signal to Noise Ratio (PSNR = ∞) and Normalized Correlation Coefficient (NCC [Formula: see text]) for lossless recovery in our python based-evaluations. Compared with other state-of-the-art chaotic and quantum-inspired encryption techniques, the proposed framework offers superior randomness, a good diffusion-confusion balance, and robustness against statistical and differential attacks. Thus, it is a promising candidate for secure image communication and high-assurance data protection in next-generation multimedia systems.