•  
  •  
 

Document Type

Article

Keywords

Color image, Confusion, Lightweight algorithm, F-Function, Gauss map

Abstract

Many lightweight algorithms, such as the tiny lightweight algorithm, have significant weaknesses, mainly due to the lack of substitution boxes, effective confusion mechanisms, or both. In today's world, enhancing encryption and secure transmission has become increasingly vital. This paper presents a newly developed lightweight algorithm for color image encryption that is based on a new symmetric block cipher structure. The method starts by transforming each pixel channel value into a 24-bit binary number. A new F-function is introduced in this block cipher to improve diffusion and confusion. Additionally, a 3D Hindmarsh-Rose model is used to generate a dynamic 6-bit S-Box in an octal format (8×8). A new approach, which is based on the Gauss map, is proposed for generating shift values, which further enhances confusion in the block cipher alongside additive and XOR operations. Python simulation experiments were conducted to analyse the security of the encryption. Tests were performed on the Lena image with a resolution of 512×512 pixels, yielding information entropy values of 7.9992 for red, 7.9990 for green, and 7.9992 for blue. The correlation coefficients were minimal, with values of red (horizontal: -0.0027, vertical: -0.0011, diagonal: 0.0019), green (horizontal: -0.0027, vertical: -0.0015, diagonal: 0.0020), and blue (horizontal: 0.0023, vertical: 0.0031, diagonal: 0.0025). Additionally, differential attack tests, including the number of pixel change rates (NPCRs) and unified average change intensities (UACIs), yielded values of 99.6048, 99.6090, and 99.6014 for the NPCRs and 33.3680, 33.4909, and 33.4099 for the UACIs across the red, green, and blue channels, respectively. The results demonstrate that the proposed algorithm provides strong encryption performance and high resistance to differential attacks.

Download

Share

COinS