Enhancing Data Security with a New Color Image Encryption Algorithm Based on 5D Chaotic System and Delta Feature for Dynamic Initialization

This research introduces an innovative color image encryption algorithm that uses a 5D chaotic system and delta feature to modify the initial conditions during key generation. It employs diffusion and confusion processes across multiple stages. Furthermore, a single pixel value in the image is altered, and the delta features are recalculated to generate new encryption keys, Subsequently, the image is encrypted once again. The number of revised keys and the similarity between the encrypted images can be measured using the Hamming distance metric. The algorithm’s effectiveness in encrypting color images is validated through simulation, testing, and comparisons with existing methods, demonstrating its effectiveness in terms of encryption quality, computational efficiency, and resistance against various attacks. The presented approach offers a promising solution for secure color image encryption, suitable for applications requiring reliable protection of sensitive visual information. The evaluation results of the algorithm, A high NPCR value (99.6099), suggest that a sizable part of the pixels has been altered during encryption, which is desirable for security. The PSNR value (7.738) shows minimal information loss during encryption, crucial to maintaining image quality. An ideal UACI value (31.131%–35.225%) suggests minimal visual artifacts and distortion in the encrypted image.