摘要
随着数字技术的飞速发展以及互联网的普及,数字技术在给人们带来方便的同时,也使得盗版变得越来越容易,盗版损失也日趋严重。数字指纹技术作为一种新型的数字版权保护技术,它是在初始数字产品中嵌入可以唯一标识用户的信息指纹,将嵌入指纹后的数字产品分发给用户,当发现盗版产品时,提取出违法指纹,使用检测算法实现盗版追踪,从而使盗版者不敢将自己所拥有的拷贝用于盗版,起到版权保护的作用。数字指纹技术的研究具有重要的科学意义和巨大的应用价值。目前的各种指纹编码几乎都存在抗合谋性不够全面、码字太长、容纳用户数量有限等缺点,离实际需要还有一段比较长的距离,实用的数字指纹应用系统还需要解决大量的关键问题,尤其是指纹系统的抗合谋性能问题。
本文对现有的几种常见的指纹编码方案进行归纳和总结,分析了它们所存在的优缺点,并对基于AND-ACC码的指纹方案(TWWL方案)和线性组合攻击进行了介绍。为了提高指纹方案的鲁棒性以及抗攻击的能力,本文做的主要工作如下：
第一,对传统的硬检测方案进行了改进。为传统硬检测算法中的检测向量设定一个上限临界值,进而捕获更多的合谋者。理论分析和实验结果表明：与传统的硬检测算法相比,改进方案提高了合谋者检测率。
第二,针对TWWL方案容易遭受线形组合攻击的特点,本文提出了一种基于图像感知特征的数字指纹方案。该方案的指纹由TWWL指纹串联构造而成,嵌入算法采用基于离散余弦变换的扩频嵌入技术,并在算法中应用人类视觉模型,利用改进的硬检测算法实现合谋者追踪。理论论证了该方案不但能抵抗平均攻击、最大值和最小值攻击,而且提高了抵抗线性组合攻击的能力。仿真实验结果也验证了本方案的有效性。
With the rapid development of digital technology and popularity of Internet, digital technology brings us great convenience on the one hand, but makes piracy easier and consequences it caused increasingly serious on the other hand. As a new digital copyright protection technology, digital fingerprinting is a technology that embeds the unique identification user's information fingerprint into original digital products, which will be distributed to users. It functions as a copyright protection by extracting illegal fingerprint and using detection algorithm to realize piracy-tracking so that those pirates don't dare to apply his copies to piracy. Researches on digital fingerprinting technology are of great scientific significance and huge application values. The existing various fingerprint codes have a series of defects such as incomplete anti-collusion, long codeword, limited number of users, which have a long way to go. Practical digital fingerprinting application system has a lot of problems to deal with, especially the problem of anti-collusion.
In this paper, we summarize several common fingerprint code schemes, analyze the merits and demerits of them, and introduce TWWL fingerprint scheme based on AND-ACC code and the linear combination collusion attack. To improve the robustness and enhance the anti-attack ability of fingerprint schemes, the main task has been done as follows:
Hard detection arithmetic has been improved which sets an upper threshold for the correlator vector of traditional hard detection arithmetic to capture more colluders. The theoretical analysis and experiment results show that this arithmetic increased probability of tracking colluders.
According to the characteristics of the TWWL scheme which is vulnerable to linear combination collusion attack, a digital fingerprinting scheme based on the image perceptive technology has been proposed. Fingerprints of the scheme are constructed in the form of TWWL's fingerprints in series. Embedding algorithm uses spreading spectrum embedding technology grounded on the spread spectrum of discrete cosine transform, and applies human visual model into arithmetic as well as utilizing improved hard detection arithmetic to track colluders. The scheme has been proved theoretically to be not only resistant to average attack, nonlinear collusion attacks such as maximum attack, minimum attack, but also enhance resisting linear combination collusion attack ability. At same time, the scheme has been proved practically to be effective by experiments.
引文
[1]Shannon C. Communication theory of secrecy systems. Bell Systems Technical Journal,1949,28(4):656-715
[2]Diffie D, Hellman M E. New directions in cryptography. IEEE Transactions on Information Theory,1976,22(6):644-654
[3]Wagner, Neal. Fingerprinting. In:Proc of the 1983 Symposium on Security and Privacy, IEEE Computer Society,1983,8-22
[4]Blakley G R, Meadows C, Purdy G B. Fingerprinting log forgiving message. In: Proc of Crypto'85, Springer, Verlag,1985,180-189
[5]Boneh D, Shaw J. Collusion-Secure fingerprinting for digital data. IEEE Transactions on Information Theory,1998,44(5):1897-1905
[6]Sebe F, Ferrer J D. Collusion-Secure and cost efficient detection of unlawful multimedia distribution. IEEE Transactions on Information Theory,2003,33(3): 382-389
[7]Barg A, Blakley G R, Kabatiansky G A. Digital fingerprinting codes:Problem statements, Constructions, Identification of Tractors. IEEE Transactions on Information Theory,2003,49(4):852-865
[8]Yacobi Y. Improved Boneh-Shaw content fingerprinting. In:Proc of RSA Conference. Heidelberg:Springer Berlin,2001,378-391
[9]Schaathum H G. The Boneh-Shaw fingerprinting scheme is better than we thought. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.64.2009&rep=rep1&type=pd f,2010-4-20
[10]李栋,卢增祥,路海明.计算机科学,2007,34(11)：224-229
[11]Proakis J G. Digital communications.4th. New York:McGraw Hill,2000, 347-362
[12]Trappe W, Wu M, Wang Z, et al. Anti-collusion fingerprinting for multimedia. IEEE Transactions on Signal Processing,2003,51(4):1069-1087
[13]Wu M, Trappe W, Wang Z J, et al. Collusion-resistant fingerprinting for multimedia. IEEE Signal Processing Magazine,2004,21(5):15-27
[14]Wang Z J, Wu M, Trappe W, et al. Anti-collusion forensics of multimedia fingerprinting using orthogonal modulation. IEEE Transactions Image Processing,2005,14(6):804-821
[15]He S, Wu M. Joint coding and embedding techniques for multimedia fingerprinting. IEEE Transactions Information Forensic Security,2006,1(2): 231-247
[16]Wu Y D. Linear combination collusion attacks and its application on an anti-collusion fingerprinting. In:IEEE Conf. on Acoustics, Speech, and Signal Processing,2005,1069-1087
[17]Seol J M, Kim S W. A scalable AND-ACC fingerprinting scheme for practical contents distribution. In:Proc of the SPIE,2005,1312-1318
[18]Wu Y D, Zhao Z G. Collusion attack to a scalable AND-ACC fingerprinting scheme. In:Proc of the SPIE,2007,1-8
[19]He S, Wu M. Collusion-resistant video fingerprinting for large user group. IEEE Transactions On Information Forensic and Security,2007,2(4):697-709
[20]王丽娜,张焕国.信息隐藏技术与应用.武汉：武汉大学出版社,2003,15-43
[21]Voloshynovskiy S, Pereira S, Iquise V, et al. Attack modeling towards a second generation watermarking benchmark. Signal Processing,2001,81(6):1177-1214
[22]季智,戴旭初.数字水印攻击技术及其对策分析.测控技术,2005,24(5)：14-17
[23]李春茹,王保保.数字水印稳健性攻击及相应策略.计算机仿真,2005,22(4)：82-85
[24]Hans G S. Novel attacks on spread-spectrum fingerprinting. EURASIP Journal on Information Security,2008,2008(2008):115-130
[25]Zhao H, Wu M, June W Z, et al. Nonlinear collusion attacks on independent fingerprints for multimedia.In:IEEE Computer Society,2003,613-616
[26]杨义先,钮心忻.多媒体信息伪装综论.通信学报,2002,23(5)：32-38
[27]Cox I, Bloom J, Miller M. Digital watermarking:Principles&Practice. San Mateo, CA:Morgan Kaufman,2001,1-10
[28]Podilchuk C, Zeng W. Image adaptive watermarking using visual models. IEEE J.SelectedAreas Commun,1998,16(4):525-538
[29]Cox I, Kilian J, Leighton E, et al. Secure spread spectrum watermarking for multimedia. IEEE Transactions Image Processing,1997,6(12):1673-1687
[30]孙圣和,陆哲明,牛夏牧等.数字水印技术与应用.北京：科学出版社,2004,127-136
[31]钮心忻.信息隐藏与数字水印.北京：北京邮电大学出版社,2004,80-86
[32]丁仲昌.抗共谋攻击数字指纹系统的研究：[南京理工大学硕士学位论文].南京：南京理工大学,2009,38-49
[33]Kilian J, Leighton T, Matheson L, et al. Resistance of digital watermarks to collusive attacks. In:Proc.IEEE Int. Syrup.Inform, Theory,1998,1-23
[34]Moulin P, Briassouli A. The gaussian fingerprinting game. In:Conference on Information Sciences and Systems,CISS'02,2002,20-22
[35]李敏.抗合谋数字指纹技术研究：[合肥工业大学硕士学位论文].合肥：合肥工业大学,2009,9-12
[36]Lofvenberg J, Lindkvist T. A general description of pirate strategies in a binary fingerprinting system. Report LITH·ISY-R-2259, ISSN,2008,1400-1437
[37]韩毅娜,尹忠海,简剑锋等.基于数字指纹的叛逆者追踪技术.空军工程大学学报,2006,7(4)：60-63
[38]王睿,方勇.基于不变矩的Copy-Move型篡改图像盲检测方法.计算机图象图形学报,2008,13(10)：1938-1941
[39]Li Q L, Wang X M, Li Y S, et al. Construction of anti-collusion codes based on cover-free families. In:IEEE Computer Society,2009,362-365
[40]卢开澄,卢华明.组合数学.北京：清华大学出版社,2002,374-382
[41]Trapped W, Wu M, Liu K J. Anti-collusion fingerprinting for multimedia technical research report. TR2002-1 7 ISR of the University of Maryland,2002, 8-12
[42]Colbourn C J, Dinitz J H. The CRC handbook of combinatorial designs. Boca Raton, FL:CRC Press,1996,25-124
[43]Dinitz J H, Stinson D R. Contemporary design theory:A Collection of Surveys, New York:Wiley,1992,371-395
[44]章毓晋.图像处理和分析.北京：清华大学出版社,1999,64-66
[45]Barlow H B. Dark, Light A. Psychophysics. Handbook of Sensory Physiology, 1972,60(3):526-530
[46]Barni M, Bartolini F, Piva A. Improved wavelet based watermarking through pixel wise masking. IEEE transactions on image processing,2001,10(5): 783-791
[47]Watson A B. DCT Quantization matrices optimized for individual images. In: Proc of the SPIE, HumVision, Visual Processing,1993,202-216
[48]Ahumada A J, Peterson H A. Luminance-model-based DCT quantization for color image compression. In:Proc of the SPIE,1992,365-374
