基于五维超混沌系统与Arnold变换的图像加密算法  

Image Encryption Algorithm Based on Five-dimensional Hyperchaotic System and Arnold Transform

作　　者：徐嘉诚 汪涛 XU Jiacheng;WANG Tao(College of Intelligent Systems Science and Engineering,Hubei Minzu University,Enshi 445000,China)

机构地区：[1]湖北民族大学智能科学与工程学院,湖北恩施445000

出　　处：《湖北民族大学学报（自然科学版）》2024年第4期486-493,550,共9页Journal of Hubei Minzu University:Natural Science Edition

摘　　要：针对低维混沌系统在进行图像加密时因结构简单而出现密钥空间小、复杂度和安全度较低等问题,提出了基于五维超混沌系统与Arnold变换的图像加密算法。该算法利用混沌系统的伪随机性与初值敏感性等特点,对混沌系统设置1个初始值增量,每次循环让混沌系统产生不同的混沌伪随机序列,根据当次循环产生的混沌伪随机序列决定Arnold变换规则和异或矩阵,对原始图像完成置乱和扩散操作,多次循环加密得到加密图像。理论分析和实验结果表明,该算法密钥空间远大于2100且密钥敏感性高,有一定的抗剪切能力和抗噪声攻击能力。加密后图像信息熵接近于理想值,相邻像素相关性大幅减小。该算法具有良好的加密性能及抗攻击能力,为图像加密提供了可行的选项。To address the problems of small key space,lower complexity and security of low-dimensional chaotic systems due to their simple structure when performing image encryption,an image encryption algorithm based on a five-dimensional hyperchaotic system with Arnold transform was proposed.Using the pseudo-randomness and initial value sensitivity of the chaotic system,the algorithm set one initial value increment for the chaotic system,and each cycle allowed the chaotic system to generate different chaotic pseudo-random sequences,which decided the Arnold transform rule and the dissimilarity matrix according to the chaotic pseudo-random sequences generated by the current cycle,and finally completed the scrambling and diffusion operations on the original image,and then encrypted it to obtain the encrypted image by several cycles.The theoretical analysis and experimental results showed that the key space of the algorithm was far larger than 2100 and the key sensitivity was high,which had some resistance to shear and noise attacks.The information entropy of the encrypted image was close to the ideal value and the correlation of adjacent pixels was significantly reduced.The algorithm has good encryption performance and attack resistance and provides a viable option for image encryption.

关 键 词：五维超混沌系统 ARNOLD变换 图像加密 混沌伪随机序列 加密性能 抗攻击能力 

分 类 号：TP309.7[自动化与计算机技术—计算机系统结构]