부산대학교 도서관

With the increasing use of digital images, there is a growing need for secure and efficient image encryption algorithms to ensure their confidentiality during transmission. In this paper, we propose an image encryption scheme for grayscale and color images that is based on primitive polynomial and Kronecker product of two invertible Suslin matrices. In the proposed scheme, primitive polynomial to map the pixel values of plain image to the elements of finite field is used, which results in pixel permutation. Then, the shuffled pixels are further diffused using a matrix obtained from the Kronecker product of two invertible Suslin matrices. To enhance security, the proposed scheme incorporates the logistic chaotic map. Further, we add authentication with the help of digital signature. Finally, the proposed scheme is implemented using Python and analyzed using various security parameters. The proposed methodology outperforms existing techniques in several critical aspects, including resistance to noise and data loss attacks, key space size, time complexity, correlation coefficient of adjacent pixels and entropy. Notably, the proposed scheme showcases a significant improvement in performance with PSNR values approaching ideal levels for noisy images and a vast 256-bit key space, bolstering security. These advancements along with the disruption of pixel correlations and increased randomness collectively makes encryption scheme a promising and secure choice for safeguarding grayscale and color images during transmission.