A Secure Color Image Encryption Method Using Genetic Algorithm–Based Pixel Permutation and Multimap Chaotic Diffusion
Rami Sihwail, Mariam Al Ghamri, Khairul Akram Zainol Ariffin · Applied Computational Intelligence and Soft Computing · 2026
Images transmitted across open networks, such as the Internet, are inherently vulnerable to eavesdropping, unauthorized access, and data tampering. Sensitive image categories, including medical, military, biometric, and legal images, often require a higher level of protection than public protocols provide. Moreover, the large size, high redundancy, and strong spatial correlation of images make it difficult to apply traditional cryptographic techniques, such as the Advanced Encryption Standard (AES) or the Data Encryption Standard (DES), which are computationally expensive for large‐scale image data. To address this problem, this paper proposes a novel, computationally efficient color image encryption method that combines genetic algorithm (GA)–based pixel permutation with multimap chaotic diffusion. The proposed method uses GA to obtain an optimal pixel permutation that eliminates spatial correlations and strengthens confusion, while diffusion based on the Logistic, Tent, and Sine chaotic maps modifies pixel values with high sensitivity to the encryption key. Evaluated on six standard color test images of varying sizes, the method achieved near‐ideal entropy of up to 7.9996, adjacent pixel correlation reduced to near‐zero values as low as −0.0050, and strong resistance to differential attacks with NPCR above 99.59% and UACI close to 33.33%. Security analysis further confirms a key space exceeding 10^84, high key sensitivity, and strong robustness against statistical, differential, and brute‐force attacks. These results show that the proposed method provides an effective and efficient solution for securing color image transmission in multimedia communication systems.