LUC Image Encryption Enhancement Using a Hyper-Chaotic System and DNA

Abstract

The digital image is considered one of the most heavily attacked and targeted data types because of its various applications and sensitive visual data. Many security techniques have been introduced to cover various image security needs, such as watermarking and steganography for intellectual property protection, authentication, and cryptography techniques for confidentiality and secrecy. In this paper, we present a technique that introduces enhancements to encryption quality and employs chaos theory to augment system robustness. The approach involves the generation of pseudo-random sequences through the utilization of the four-dimensional Lorenz system. These sequences serve a dual purpose: firstly, they facilitate image preprocessing by applying a scrambling and diffusion technique to eliminate intrinsic relationships between adjacent pixels. Secondly, a DNA-based codebook method is incorporated to further elevate encryption quality during the final step. This entails the utilization of the LUC public encryption algorithm, along with one of the generated pseudo-random sequences, as the encryption keys for safeguarding the image. The findings regarding encryption quality and the perceptibility of the resulting image yield promising outcomes. In terms of encryption quality, notably high values were attained. In regard to entropy, the achieved entropy for the encrypted images exceeded 7.9. Additionally, the range of PSNR between the plain and encrypted images remained below 8.7.