一种基于时变符号动力系统的H.265视频加密算法与实现

发布时间：2018-06-19 16:25

本文选题：符号动力系统 + 混沌加密　；参考：《深圳大学》2017年硕士论文

【摘要】：随着信息技术与计算机技术的高速发展,人们对高清数字视频信息的需求越来越广泛,而视频高清化的广泛需要也逐步提高了人们对视频压缩性能的要求。为此,在2010年4月,由VCEG和MPEG再次组建的视频编码联合组(JCT-VC)制定了新一代视频压缩标准(H.265)。不过,H.265标准并未考虑如何防止视频数据遭人窃取、恶意破坏和篡改等问题,因而研究一套适合这种新标准H.265的视频加密算法的问题就突显出来了。本文所研究的课题部分得到了深圳市学科布局项目“基于光编码物理层安全的光通信关键技术研究”(JCYJ20160328145357990)的资助,是以H.265标准的加密算法为研究重点,结合混沌序列密码和H.265压缩标准的特点,设计了一种新的基于混沌序列密码的视频加密方案。该方案首先利用时变符号动力系统设计了伪随机性能良好的混沌序列,之后在改进熵编码中原有算术编码方案后,利用混沌伪随机序列和改进的算术编码设计了一种适合H.265特性的视频加密方案,最后通过理论分析和仿真说明了该视频加密算法的良好效果。总体来说,本文工作的创新点包括:1)在理论上严格证明了一种时变广义符号动力系统的混沌性,并结合寄存器序列与混沌序列,提出了伪随机序列的一种新的产生方法,之后利用单比特频数检测、扑克检测以及自相关检测等常见伪随机性检测方法对所产生的序列进行性能检测,最后利用仿真实验验证了该序列的随机性能优良。这为混沌序列密码中关键的密钥流序列的设计打下了基础。2)利用新的时变符号混沌系统,设计一种新的混沌序列加密算法,将其应用于数字图像加密中,通过仿真实验检验了良好的加密效果,并说明了该图像加密算法的密钥空间大、密钥敏感性强、抗统计攻击性好等优点。3)基于新标准H.265中原有算术编码的特点,提出了一种算术编码的改进方案,并结合新的时变符号混沌系统设计了一种新的视频加密方案,之后通过理论分析了该方案的合理性,最后利用HM15.0仿真平台验证了该方案具有良好的加密效果,并说明了该视频加密算法的压缩效率高、压缩比和视频数据格式不变等优点。
[Abstract]:With the rapid development of information technology and computer technology, the demand for high definition digital video information is becoming more and more widespread, and the wide demand for video high-definition has gradually improved the demand for video compression performance. In April 2010, the video coding joint group (JCT-VC), which was set up by VCEG and MPEG, made a new generation of video. Frequency compression standard (H.265). However, the H.265 standard does not consider how to prevent video data from being stolen, malicious destruction and tampering. Therefore, the study of a set of video encryption algorithms suitable for this new standard H.265 is highlighted. The subject of this paper has been studied in the Shenzhen discipline layout project based on optical coding. The research on the key technology of physical layer safety of optical communication (JCYJ20160328145357990) is to design a new video encryption scheme based on chaotic sequence cipher based on the characteristics of H.265 standard encryption algorithm, combining the characteristics of chaotic sequence cipher and H.265 compression standard. The scheme first uses time-varying symbol power system. After calculating the chaotic sequence with good pseudo random performance, after improving the entropy coding, a video encryption scheme suitable for H.265 characteristics is designed by using chaotic pseudo random sequence and improved arithmetic coding. Finally, the good effect of the video encryption algorithm is illustrated by theoretical analysis and simulation. The innovative points of the work include: 1) in theory, the chaos of a time-varying generalized symbolic power system is strictly proved, and a new generation method of pseudo random sequence is proposed in combination with the register sequence and the chaotic sequence. After that, the common pseudo random detection methods, such as single bit frequency detection, poker test and autocorrelation detection, are used. The performance of the sequence is tested. Finally, the simulation experiment shows that the sequence has good random performance. This is the basic.2 for the design of key key stream sequence in chaotic sequence cipher. A new chaotic sequence encryption algorithm is designed and applied to digital image encryption by using the new time-varying symbolic chaotic system. The good encryption effect is tested by the simulation experiment, and it shows that the key space of the image encryption algorithm is large, the key sensitivity is strong and the anti statistical attack is good.3.) based on the new standard H.265 has the characteristics of arithmetic coding in the central plain, an improved case of arithmetic coding is put forward, and a new time-varying symbol chaotic system is designed. The new video encryption scheme is introduced, and the rationality of the scheme is analyzed through the theory. Finally, the HM15.0 simulation platform is used to verify the good encryption effect of the scheme, and the high compression efficiency of the video encryption algorithm, the compression ratio and the invariant video data format are also illustrated.
【学位授予单位】：深圳大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN919.81;TP309.7

【参考文献】