Optimized Image Encryption Model based on Hybridization of Chaotic Maps with Metaheuristic OOBO Algorithm
Abhinaya Srivastava, Shano Solanki · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2025
Chaotic maps have gained popularity in image encryption models due to their ability to secure images with a random key.Further, the randomness of the key is highly dependent on the initial parameter values of the chaotic map.Therefore, the literature employs metaheuristic algorithms to find the optimal parameter values for generating a completely random key.Two chaotic maps-the 1-D chaotic logistic map (CLM) and the chaotic tent map (CTM)-along with the metaheuristic algorithm known as one-to-one-based optimization (OOBO) were used to create an optimized image encryption (IE) model in this study.We utilize the 1-D CLM algorithm for random key generation, while the 1-D CTM algorithm generates the shuffling sequence for image shuffling.We use the metaheuristic OOBO algorithm in both chaotic maps to generate completely random keys and random shuffling sequences based on the objective function.In this research, we design the objective function using security parameters such as entropy (E) and correlation coefficient (CC).After determining the optimal parameter values for chaotic maps, we perform an exclusive-OR operation between the random key and the secret image.After that, shuffling of the secret image is done based on the shuffling sequence given by the 1-D CTM algorithm.We evaluate the proposed IE model through simulation using images from the primary and secondary datasets.Using various parameters, we compare the visual quality of the encrypted image with the secret image in this evaluation.The results illustrate that the proposed IE model outperforms in terms of entropy and correlation coefficients and provides better encryption than the existing IE models.The proposed model achieves an entropy of 7.99648 for the cameraman, 7.99688 for Elaine, and 7.99704 for the pepper image.Further, a correlation coefficient of 0.00021 for the cameraman, -0.00012 for Elaine, and -0.00161 for the pepper image.