Enhancing colour image encryption through parameters optimization of memristive hyperchaotic system with CPSO algorithm and LSAIM
Biniyam Ayele Belete, Demissie Jobir Gelmecha, Ram Sewak Singh · The Imaging Science Journal · 2025
Particle Swarm Optimization (PSO) is a simple and effective algorithm for optimization. However, it often converges too early and depends heavily on proper parameter settings for good results, especially in complex or high-dimensional search spaces. Therefore, this algorithm is inappropriate for parameter optimization of hyperchaotic systems and may not provide sufficient robustness and security of colour images. This study presents a new colour image encryption based on optimizing a four-dimensional Memristor-based hyperchaotic system, in which the fourteen parameters of the hyperchaotic system are optimized by the Chaotic Particle Swarm Optimization (CPSO) method, in combination with DNA code and a new logistic sine adjusted integrated map (LSAIM). Simulation results have shown that the algorithm achieved a key space of up to 21116, providing sufficient protection against brute-force attacks, with entropy values close to the ideal (7.9994), NPCR of 99.6178%, and UACI of 33.965%, indicating robust security.