Image Encryption Model based on Chaotic Maps and Group Learning Algorithm

Abstract

Nowadays, digital images are used in several applications to communicate the information. Therefore, we employ image encryption methods to safeguard the information. Furthermore, chaotic maps have gained popularity in image encryption (IE) methods due to their improved security. Therefore, in this paper, we have proposed an optimized color IE model using the 3-D chaotic logistic map (CLM) and 1-D chaotic tent map (CTM). We employ the 3-D CLM for random key generation, while the 1-D CTM generates the shuffle indexes of the secret image. Furthermore, we incorporate a group learning (GL) algorithm to refine the initial set of parameters of the 3-D CLM. Moreover, in this research, a multi-objective is developed for the GL algorithm using various security parameters, namely, entropy, SSIM, and CC. In the encryption process, different planes of the secret color image are extracted, and performed operation with the random key is generated with 3-D CLM algorithm and shuffled the image based on the shuffling index is given by 1-D CTM. Next, the proposed IE model is evaluated for several images based on visual quality analysis and analysing the encrypted image (EI) characteristics with secret image (SI). Thus, the encrypted image histograms are completely distributed, as anticipated by the encryption model. In addition, the proposed IE model met the desired security benchmark parameters, namely, CC, SSIM, PSNR, and entropy in the range of +0.006 to −0.00231, 0 ???????? 0.011133, 8.57−23.93????????, and 7.9948−7.9971, respectively. Finally, we used the entropy and CC parameters are used to evaluate the proposed IE model over others.