A visual meaningful encryption and hiding algorithm for multiple images

Guodong Ye,Liming Guo

DOI: https://doi.org/10.1007/s11071-024-09790-7

IF: 5.741

2024-06-10

Nonlinear Dynamics

Abstract:To protect the privacy of multiple images simultaneously in the digital information era, this paper proposes a novel multi-image encryption and hiding algorithm (MiEHA) with the help of a new three-dimensional coupled chaotic model (3D-CCM), cross-plane confusion, diffusion and a three-dimensional discrete cosine transform (3D-DCT). First, to address security concerns arising from the limited performance of the current chaotic system, a new 3D-CCM is presented with a designed three-dimensional coupled map using Logistic map and Iterative chaotic map (3D-LICM). In addition, a novel key acquisition model is also developed to generate pseudo-random keystream. Next, the secret plain images are reorganized into three matrices of the same size together with the keystream. Then, the proposed algorithm implements cross-plane confusion, 3D Zigzag confusion-diffusion and cross-plane diffusion to achieve encryption for the former three matrices, resulting in a cipher image. Finally, execute a 3D-DCT transformation on a color carrier image, followed by an embedding operation of the cipher image into DCT coefficients, resulting in a carrier image containing secrets, i.e., visually meaningful cipher image. Fortunately, the proposed multi-image encryption can address the issue of requiring multiple rounds of encryption to different images. Different from noise-like cipher image, the proposed MiEHA has a better capability of encrypting and hiding secret images, with a high visual effect for carrier image before and after being embedded.

engineering, mechanical,mechanics