Enhanced chaotic active cryptography using metaheuristic optimization with image-dependent pixel displacement, bitwise operations, and chaotic key generation

Abstract With the rapid expansion of digital images, the internet of things, and smart systems, the development of secure and efficient image encryption methods has become a necessity. In the meantime, the use of chaotic systems and metaheuristic algorithms can improve the security of images against statistical attacks and brute force searches by increasing randomization and optimizing encryption parameters. In this research, a chaotic encryption approach is presented both in the stage of generating the private key and in encrypting image information with system modifications under metaheuristic algorithms. In the first stage, a private key is generated with the help of a Lorenz chaos model, which is modified with the help of the starfish optimization algorithm (SFOA) to best generate the model parameters. Then, in the image encryption stage, a combined encryption process is used, including pixel position shifting based on the Arnold Cat chaos system optimized with the goose optimization algorithm (GOA) and bitwise information processing operations for image pixels based on rotation dependent on pixel information and XOR operator with the private key. Various analyses including visual inspection, attack analysis and quantitative evaluations are used to assess the security of the system. High values of NPCR and UACI indices equal to 99.72% and 34.24%, respectively, indicate favorable resistance to differential attacks. In addition, comparison with several advanced cryptographic methods and similar metaheuristic algorithms shows that this approach provides a high level of randomness in the encrypted data by achieving entropy values of 7.9995 for 8-bit images and improves cryptographic security. The results of this cryptography show that the application of this method in real-time encryption has the complexity of chaotic encryption that attackers will not be able to decrypt the images.