【作者】 刘军；

【导师】 谢长生； 黄浩；

【作者基本信息】 华中科技大学 ， 计算机系统结构， 2009， 博士

【摘要】 随着数字多媒体技术的飞速发展，数字版权管理系统日渐引起人们的重视。而国内光盘应用市场的扩展，使得基于光盘的数字版权管理系统的研究已成为当前迫切需要解决的问题。基于光盘的数字版权管理系统中，进行版权追踪的数字水印、高效安全的数字视频加密、无状态设备的密钥管理是其三个主要研究方面。作为一种新兴的有效处理盗版追踪的版权保护方法，数字水印技术已经成为数字版权保护领域的一个研究热点。这里提出一种新的基于混沌加密的分块嵌入变换域的图像数字水印方法。将二进制水印图像信息通过混沌序列进行加密，再使用GRAY码进一步置乱弱化水印信息，然后利用人类视觉模型特性，将置乱信号自适应地加入到彩色载体图像蓝色分量离散余弦变换域的低频分量系数中。其次在此基础上进一步给出一种盲水印植入算法。针对数字水印在图像中容易破坏同步信息的问题，提出基于SIFT校正的水印提取方法，大大地提高了前面的水印算法的鲁棒性。最后还给出了针对一种空域水印植入方法中同步信息破坏的问题，以SIFT算子来校正图像的同步信息，并提高该空域水印算法的提取效率，增强了水印的鲁棒性。由于音频视频数据量大，编码解码实时性要求高，如果只是简单应用基于文本数据加密的传统加密算法会产生诸多弊端。通过深入研究了国内外视频加密算法的发展现状，分析包括选择性加密方案在内的诸多视频加密方案的优缺点，发现为了设计一个能满足光盘应用的视频加密方案的关键是从以下两个方面：如何减少加密运算量开销，同时兼顾视频关键数据的安全性。为此，针对光盘版权管理应用提出一种基于视频图像的分级安全方法。在此基础上，提出了一种新的基于X264编码视频文件的TWVEA方案(Two-way Video Encryption Algorithm)。该算法改进VEA算法选择关键数据进行加密、加密过程与视频编码相结合，这两项措施共同降低了加密运算额外开销和冗余信息量。算法实现过程中首先研究了视频中待加密的关键数据的选取，然后根据编码器的特点设计了视频加密最佳位置选取方案，最后用AES(Advanced Encryption Standard)加密算法对选取的关键数据在最佳位置进行加密，使用VEA算法对次关键数据进行加密，使用改进VEA算法对更次关键数据进行加密。此外，通过基于视频图像的分级安全方法，提出另外一类组合是快速选择性视频加密算法CA-AES-Seal组合加密方案。该加密方案在安全分级上，通过安全性和加密速度的取舍，将高安全低解密速度的AES加密算法用于选取的关键数据的加密，而较低安全高解密速度的Seal算法用次关键数据的加密，从而实现组合视频快速加密解密。最后对基于MPEG-2编码视频文件的RVEA视频加密方案进行了分析，发现基于视频图像的分级安全方法能够进一步改进RVEA视频加密方案的安全性，同时能够兼顾视频加密运算开销，并给出了两个视频加密方案的速度对比分析。根据无状态设备的特征，深入研究了国内外密钥管理算法的发展现状，首先介绍了子集覆盖算法框架，并分析了完备子集和子集差分算法的具体算法描述，重点分析了这些算法的优缺点。在此基础上，提出一种新的层次链式密钥管理方法，并给出这种层次链式密钥管理方法的详细描述和安全性分析，最后给出几种无状态设备密钥管理方法的仿真对比。最后，从光盘版权管理系统的角度出发，给出一种光盘版权保护系统的框架，给出系统版权保护中整个数据传递的密钥安全管理流程，并给出视频加密和密钥管理方法在光盘版权保护系统中具体的应用，最后提出一种基于视频水印的追踪方法，进一步完善整个光盘版权保护系统。更多还原

【Abstract】 With the development of digital multimedia technology, digital rights management has become a pressing concern for the digital content business. With the widespread applications of optical disc in China, copyright protection of multimedia content in optical disc system being sold is a key problem to be solved. The thesis studies some problems of copyright protection for optical disc system through three main respects: digital watermark, digital video encryption, key management for stateless receivers.As a new effective technological means to the copyright protection for copy tracing, the digital watermarking technology has already become a popular topic in the field of digital copyright protection. First, this thesis presents a new robust watermarking scheme using GRAY code and chaotic encryption for color image. A binary watermark image is permutated using sequence numbers generated by a secret key in a spatiotemporal chaos system. Then, the binary image watermark as a unit was encoded by Gray code, and then adaptively embedded into low frequency components of the original color image’s blue channel in discrete cosine transform domain. Second, an improved robust blind watermarking scheme using chaotic encryption for color image is proposed. Third, these watermarking schemes have some technical limitation in robustness against geometric attacks, especially in robustness against global affine transformation, such as rotation, scaling and translation. Because these desynchronization attacks, which induce synchronization errors between the original and the extracted watermark during the detection process, are very difficult to tackle, this thesis proposes a new robust watermarking extract scheme for color image using scale invariant features transform (SIFT) image correction. Experimental results show that the proposed scheme is robust and secure against a wide range of image processing operations such as global affine transformation, filtering and lossy compression. Fourth, the same method that watermarking extract scheme for color image using scale invariant features transform (SIFT) image correction is used to improve another watermarking detected algorithm for color image based on a block probability in spatial domain.Although traditional text encryption approaches can be used to encrypt video data, it has many disadvantages in encrypting mass data at a high real time capability in multimedia application. We must develop a new way to solve the problem of digital video encryption and decipher. Then the selective encryption algorithm becomes the new hotspot of video encryption. The key to develop a video encryption algorithm to reach the real-time requirement and security quality is to make a good balance between security and the excessive cost of the encryption arithmetic operation. The security quality and the operation time of the algorithm will fall if we reduce the complexity of the encryption; conversely, the security quality and the operation time will increase with the decrease of the complexity of the algorithm. A new grade method for video security quality is proposed, and it distinguishes four ranks: important, sub important, common and non requisite. Then a two-way selective encryption algorithm (TWVEA) for MPEG-4 AVC video based on this video security grade method is proposed to compromise the security and speed in the process of encryption. TWVEA improves VEA algorithm and adopts key video data to encrypt during the encoding progress to reduce the excessive operation spending; And three different encryption algorithm encrypt the video data in different security level. These encryption methods include AES, VEA and improved VEA, which are used to encrypt the three areas in one image of video data. This selective encryption of MPEG video streams reduces the required data of encryption, saves in complexity and facilitates new system functionality. After that, another fast selective encryption algorithm named CA-AES-Seal is proposed. It’s a new algorithm that combines AES encryption method and Seal encryption method. To make a trade-off between security and encryption cost, AES method is used to protect the core key video data to receive high security quality and Seal method is used to encrypt the sub important video data to realize fast encryption algorithm. Finally, the video encryption scheme named RVEA, which is based on the MPEG-2 video files, is analyzed. And the new grade method for video security quality is found to further can enhance security of RVEA. Both comparative analysis of computing overhead of these video encryption schemes are given.When a legitimate receiver with its initial set-up can obtain the current group session key only from the current transmission without the history of past transmissions, it is named a stateless receiver. Broadcast encryption schemes are one-to-many communication methods in which a message sender can broadcast an encrypted data to a group of users over an insecure channel so that only legitimate receivers can decrypt it. And how to efficiently exclude illegal (revoked) users from a privileged set is the task of the stateless broadcast encryption schemes. So the stateless broadcast encryption schemes need to ensure that only legal users decrypt an encrypted broadcast message. By studying the development trend of broadcast encryption schemes at home and abroad, first of all this thesis introduces the Subset-Cover algorithm framework, and analyzes the complete subset algorithm and Subset-Difference algorithm, focusing on analysis of the advantages and disadvantages of these algorithms. On this basis, a new Layer-Chain key management method is proposed, and the Layer-Chain key management method is described in detail. The comparison of the stateless broadcast encryption schemes shows that the Layer-Chain key management method has a reasonably low user storage and computation overhead.Then, an optical disc copyright protection system framework is given in detail, and the workflow of the safety key management processes shows that video encryption algorithm and Layer-Chain key management method are suitable for the optical disc copyright protection system. Finally a new copy tracing scheme based on video watermark is proposed, which further improves the entire optical disc copyright protection system.更多还原