基于分段Logistic混沌映射的自适应图像加密算法的研究

发布时间：2018-05-05 01:02

本文选题：混沌系统 + 自适应　；参考：《安徽大学》2017年硕士论文

【摘要】：由于Internet的蓬勃发展,多种多样的图形和图像等多媒体信息在互联网络上传播。然而,据报道世界上每20秒就有一例黑客入侵事件发生,许多重要的数据在信息传送过程中破损或被窃取。怎样准确完整地传送所有信息是目前网络信息中一个急需解决的难题。数字图像在多媒体信息中占据着重要的位置,对图像进行加密是目前可靠传输的主要方式。由于混沌系统的内在随机性和初值敏感性与已有的古典密码体制的扩散性质相契合,因此被广泛应用到图像加密中。Logistic混沌映射具有复杂的动力学行为,其所产生的混沌密钥序列具有良好的随机性,可以改变图像的灰度值,而自适应加密算法是以图像自身的像素对图像进行排列置乱的加密算法,具有运行速度快、置乱效果好的优点,因此较多应用于图像加密中去。本文在深入研究基于混沌理论的图像加密的基础上,结合自适应技术,对混沌映射的图像加密进行了改进。本文的主要工作为以下几个方面。1.采用了一种两段式Logistic混沌映射对图像进行加密。本文在研究多种一维和高维的混沌系统的基础上,选择一种对一维Logistic映射进行改进的两段式Logistic混沌映射,并应用在图像加密中。该两段式不仅比原有Logistic混沌映射更早进入混沌状态,而且在很大参数范围内都达到满映射状态,增大了密钥空间,增强了一维Logistic混沌映射的抗攻击性能力。并通过统计分析、相邻像素的相关性分析等验证了该改进加密算法的安全性。2.设计了一种基于两段式Logistic混沌映射的自适应图像加密算法。自适应算法自提出以来就受到广大学者的关注,它改进了一般排列加密与密钥无关的缺点,但是排列加密只改变图像像素的位置,像素之间的相关性没有改变。本文把自适应加密算法的排列加密与像素替代相结合,设计了一种基于两段式Logistic映射的自适应图像加密算法,在对图像分块后,按照图像加密顺序,用第一个子块图像的像素异或值对下面子块图像进行加密。在分段Logistic混沌映射的基础上,通过Logistic映射产生随机序列并转换成二进制密钥流,采用主密钥流对图像随机选择加密顺序,子密钥流选择对子块的像素排列与像素替代的混合加密模式。通过实验仿真和差分性分析、直方图分析、相关性分析验证了算法的安全性。
[Abstract]:With the rapid development of Internet, a variety of multimedia information, such as graphics and images, is spread over the Internet. However, it is reported that there is a hacking incident every 20 seconds in the world, and many important data are broken or stolen in the process of information transmission. How to transmit all information accurately and completely is an urgent problem in network information. Digital image plays an important role in multimedia information. Encryption of image is the main way of reliable transmission. Because the inherent randomness and initial value sensitivity of chaotic systems coincide with the diffusion properties of the existing classical cryptosystems, it is widely used in image encryption. Logistic chaotic mapping has complex dynamic behavior. The chaotic key sequence produced by the algorithm has good randomness and can change the gray value of the image. The adaptive encryption algorithm is an encryption algorithm that arranges and scrambles the image with its own pixels. The effect of scrambling is good, so it is more used in image encryption. Based on the research of image encryption based on chaos theory and adaptive technology, this paper improves the image encryption of chaotic map. The main work of this paper is as follows. A two-segment Logistic chaotic map is used to encrypt the image. In this paper, based on the study of many kinds of chaotic systems of one dimension and high dimension, we choose a two-segment Logistic chaotic map, which can improve the one-dimensional Logistic map, and apply it to image encryption. The two segments not only entered the chaotic state earlier than the original Logistic chaotic map, but also reached the full mapping state in a wide range of parameters, which increased the key space and enhanced the anti-attack ability of the one-dimensional Logistic chaotic map. The security of the improved encryption algorithm is verified by statistical analysis, correlation analysis of adjacent pixels and so on. An adaptive image encryption algorithm based on two-segment Logistic chaotic mapping is designed. The adaptive algorithm has been concerned by many scholars since it was put forward. It improves the shortcomings of general permutation encryption which is independent of the key, but the permutation encryption only changes the location of image pixels, and the correlation between pixels is not changed. This paper combines the arrangement encryption of adaptive encryption algorithm with pixel substitution, and designs an adaptive image encryption algorithm based on two-segment Logistic mapping. Encrypt the subblock image below with the pixel XOR value of the first subblock image. On the basis of piecewise Logistic chaotic mapping, random sequence is generated by Logistic mapping and converted into binary key stream. The subkey stream selects the pixel arrangement of the sub-block and the mixed encryption mode instead of the pixel. The security of the algorithm is verified by experimental simulation, differential analysis, histogram analysis and correlation analysis.
【学位授予单位】：安徽大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP309.7;TP391.41

【参考文献】