基于CNN高维超混沌系统图像加密算法  

An image encryption algorithm based on CNN hyperchaotic system

作　　者：张琴 林达 余亮 ZHANG Qin, LIN Da, YU Liang(1. College of Automation and Information Engineering, College of Physics and Electronics Engineering, Zigong 643000, China ; 2. Siehuan University of Science ＆ Engineering, College of Physics and Electronics Engineering,Zigong 643000, China)

机构地区：[1]四川理工学院自动化与信息工程学院,四川自贡643000 [2]四川理工学院物理与电子工程学院,四川自贡643000

出　　处：《西安邮电大学学报》2018年第2期32-39,共8页Journal of Xi’an University of Posts and Telecommunications

基　　金：国家自然科学基金项目(61640223);人工智能四川省重点实验室开放基金(2016RZJ02)

摘　　要：基于高维细胞神经网络(CNN)混沌系统,提出了一种灰度图像加密算法。首先,利用仿真数值方法,分析了CNN系统的复杂的动力学特性和初值敏感性等特点,并利用其系统特点设计了一种新的伪随机序列发生器。然后,利用伪随机序列发生器生成的伪随机序列密钥对待加密图像进行置乱,再进行横向和纵向扩散,实现对图像进行加密。最后,通过实验验证了算法的有效性,测试了算法得到安全性,结果表明该算法具有较好的安全性、鲁棒性等优点。A gray image encryption algorithm based on high dimensional cellular neural network hyperchaotic system is proposed.In this algorithm,firstly the complex dynamic characteristics and initial value sensitivity of CNN system are analyzed by using simulation numerical methods,and a new pseudo-random sequence generator is designed based on its system characteristics.Secondly,the image to be encrypted is scrambled using the pseudo-random sequence key generated by the pseudo-random sequence generator,and then horizontal and vertical diffusion is performed to implement the image encryption.Finally,the effectiveness of the algorithm is verified by experiments,and the security of the algorithm is tested.The results show that the algorithm has good security and robustness.

关 键 词：高维超混沌 细胞神经网络 伪随机序列发生器 图像加密 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]