基于遗传算法的数字水印技术研究
|
摘要
为了减小水印嵌入对图像视觉效果的影响,本文在空间域利用遗传算法研究水印的最佳嵌入问题,着重考虑了遗传算法的决策变量编码、算子选择和参数设计等相关内容,主要工作如下:
1.介绍了数字水印技术的基本概念和原理,综述了数字水印技术的发展现状,分析了现有嵌入算法中的不足。
2.提出了一种基于遗传算法的数字水印空域优化嵌入技术,给出了水印嵌入算法和提取算法。
3.分析了决策变量的基因型和表现型之间一一对应关系,提出了应用嵌入点位置值作为决策变量基因型的实数编码算法,与传统的二进制编码相比,提高了运算精度和效率,减少了计算复杂度。
4.分析了遗传运算中的三个遗传算子、四个运行参数和一些必须的约束条件之间的相互关系,给出了算子和参数选择的最优方法。
5.分别对应用普通算法和遗传算法嵌入水印后的图像进行了抗攻击性和不可见性的对比实验,表明在抗攻击性要求一致的条件下,遗传算法的使用提高了图像的不可见性。
Digital watermarking technique has become a new network safety technique in recent years, some digital information, such as user IDs, products symbols, significant letters and so on, can be embedded into the multimedia data in order to protect data, which do not affect the value of original contents and can not be seen and noticed on the images. The watermarks can be extracted as often as it is needed in order to achieve copyright protection.
While using the algorithms of the digital watermarking technique applied in the images, generally, people use human visual model theory to seek for the watermark embedded method that has the smallest visual damage on the original data to avoid damaging the visual quality of the original images. In general, the traditional methods can not achieve the optimal embedding. In order to find a method that will not affect the visual effects basically, the spatial domain algorithm is applied in this thesis and the genetic algorithm is brought in to seek for an optimal watermarking image embedding plan.
Choosing appropriate parameters in the watermarking system is a problem of looking for optimal solutions or sub-optimal solutions in a big solution space. However, the generic algorithm is an optimized algorithm with stronger robustness including two distinct characteristics, the parallelism property and global searching property. The generic algorithm is quite good at solving global optimizing problems and the optimizing target of the generic algorithm in the digital watermark is to find suitable embedding positions. The main idea of the target is using the generic algorithm to optimize the objective function composed by the invisible property.
The digital watermarking technique based on the generic algorithm used the genetic algorithm to find the best point of the watermark embedding in time domain and embedded watermark directly, then the computing efficiency and image quality. in accordance with the requirements of the fitness are improved.
This thesis is mainly focusing on the application of the genetic algorithm in the process of the digital watermarking optimal embedding, and includes six research tasks shown below:
(1) In this thesis, the researching background and the development are described generally, the basic characteristics, the principles, the classification of the watermark, the applied fields and the current researching situation home and abroad are introduced, the basic structure of the digital watermarking system is generalized and the generative algorithms, the embedding algorithms and the testing algorithms of the watermarks are introduced in detail. Moreover, the research achievements and the developments in recent years home and abroad,, several current problems under disputation and a few typical embedding algorithms in the watermarking embedding technique are summarized systematically.
(2) The biologic basis, the basic principle, the technological process and operating process of the generic algorithm are introduced and the features, the development and the application of the algorithm are proposed in detail. Finally, comparing the generic algorithm and the traditional algorithms, including the comparison of the generic algorithm and the heuristic algorithm, the comparison of the generic algorithm and the climbing algorithm, the comparison of the generic algorithm and the exhaust algorithm and the comparison of the generic algorithm and the blind random algorithm, the thesis obtains the superiority of the generic algorithm in practice.
(3) A digital graphic watermarking algorithm based on the generic algorithm is proposed. The basic implementation techniques including the encoding method, the fitness function, the selecting of the generic operators, the processing of the working parameters and the constrained conditions of the generic algorithm are introduced. Finally, the thesis describes the embedding algorithm and the extracting algorithm in detail.
(4) In the process of the generic computed encoding, since the value of the decision variable is large compared with normal ones in this thesis, normal binary encoding is not predominant enough. Creatively, the decimal encoding method is applied in the thesis that can not only improve the operating accuracy of the generic algorithm, but reduce the calculated complexity and increase the operating efficiency.
(5) In the generic algorithm, there are three generic operators: selection operator, crossover operator and mutation operator, and there are gambling site selection, ranking selection and the best individual choice in the selection operator; single-point crossover, the two point crossover and multi-point crossover in the crossover operator; basic bit mutation, Gaussian mutation and uniform mutation in the mutation operator. In this paper, the operator of each study and experimental verification of each operator selected the best algorithms.
(6) In the generic algorithm, there are four operating parameters. As setting of the parameters and processing of the constrained conditions are associated with the accuracy, the reliability, calculating time and many other factors of the generic algorithm, and they may even influence the result quality and the system performance. Therefore, by several experiments, the thesis gives the optimal operating method.
(7) In the application of genetic algorithm, it is necessary to deal with constraints, which constitute in three types of methods: the search space limited to law, a viable solution and the penalty function method. But there has yet to find a way to deal with the general method of constraint conditions. Therefore, in terms of processing constraints, according to different application, constraints characteristics, and the computing ability we choose different approaches.
(8) Through the simulating experiments on the computer, the results show that the application of the generic algorithm can optimize the watermarking embedding and improve the graphic quality after embedding the watermarks greatly Innovative work is as follows:
(1) In the time domain, applying the generic algorithm to find optimal watermark embedding points to embed watermarks and the peak value signal to noise ratio to be the fitness function could improve the graphic quality embedded by the watermarks. Moreover, the algorithms are simple compared with previous ones, and the real-time property becomes much better.
(2) In the generic encoding, the encoding using real numbers is applied, which means taking the value representing the position of the graphic pixel points to be the decision variables. Compared with binary encoding, this method can avoid the conflicts of the embedding points caused by the crossover operation and mutation operation, increase the mathematical precision, improve the computing complexity and enhance the mathematical efficiency.
(3) The selection operators used in this thesis are to combine the Roulette Wheel Selection, the sorting selection and the optimal individual retaining, which means that to combine the proportional selection and the elitist genetic algorithm. This method not only can ensure the individuals with higher fitness value can be chosen as a priority, but can eliminate the individuals with much lower fitness value.
1.介绍了数字水印技术的基本概念和原理,综述了数字水印技术的发展现状,分析了现有嵌入算法中的不足。
2.提出了一种基于遗传算法的数字水印空域优化嵌入技术,给出了水印嵌入算法和提取算法。
3.分析了决策变量的基因型和表现型之间一一对应关系,提出了应用嵌入点位置值作为决策变量基因型的实数编码算法,与传统的二进制编码相比,提高了运算精度和效率,减少了计算复杂度。
4.分析了遗传运算中的三个遗传算子、四个运行参数和一些必须的约束条件之间的相互关系,给出了算子和参数选择的最优方法。
5.分别对应用普通算法和遗传算法嵌入水印后的图像进行了抗攻击性和不可见性的对比实验,表明在抗攻击性要求一致的条件下,遗传算法的使用提高了图像的不可见性。
Digital watermarking technique has become a new network safety technique in recent years, some digital information, such as user IDs, products symbols, significant letters and so on, can be embedded into the multimedia data in order to protect data, which do not affect the value of original contents and can not be seen and noticed on the images. The watermarks can be extracted as often as it is needed in order to achieve copyright protection.
While using the algorithms of the digital watermarking technique applied in the images, generally, people use human visual model theory to seek for the watermark embedded method that has the smallest visual damage on the original data to avoid damaging the visual quality of the original images. In general, the traditional methods can not achieve the optimal embedding. In order to find a method that will not affect the visual effects basically, the spatial domain algorithm is applied in this thesis and the genetic algorithm is brought in to seek for an optimal watermarking image embedding plan.
Choosing appropriate parameters in the watermarking system is a problem of looking for optimal solutions or sub-optimal solutions in a big solution space. However, the generic algorithm is an optimized algorithm with stronger robustness including two distinct characteristics, the parallelism property and global searching property. The generic algorithm is quite good at solving global optimizing problems and the optimizing target of the generic algorithm in the digital watermark is to find suitable embedding positions. The main idea of the target is using the generic algorithm to optimize the objective function composed by the invisible property.
The digital watermarking technique based on the generic algorithm used the genetic algorithm to find the best point of the watermark embedding in time domain and embedded watermark directly, then the computing efficiency and image quality. in accordance with the requirements of the fitness are improved.
This thesis is mainly focusing on the application of the genetic algorithm in the process of the digital watermarking optimal embedding, and includes six research tasks shown below:
(1) In this thesis, the researching background and the development are described generally, the basic characteristics, the principles, the classification of the watermark, the applied fields and the current researching situation home and abroad are introduced, the basic structure of the digital watermarking system is generalized and the generative algorithms, the embedding algorithms and the testing algorithms of the watermarks are introduced in detail. Moreover, the research achievements and the developments in recent years home and abroad,, several current problems under disputation and a few typical embedding algorithms in the watermarking embedding technique are summarized systematically.
(2) The biologic basis, the basic principle, the technological process and operating process of the generic algorithm are introduced and the features, the development and the application of the algorithm are proposed in detail. Finally, comparing the generic algorithm and the traditional algorithms, including the comparison of the generic algorithm and the heuristic algorithm, the comparison of the generic algorithm and the climbing algorithm, the comparison of the generic algorithm and the exhaust algorithm and the comparison of the generic algorithm and the blind random algorithm, the thesis obtains the superiority of the generic algorithm in practice.
(3) A digital graphic watermarking algorithm based on the generic algorithm is proposed. The basic implementation techniques including the encoding method, the fitness function, the selecting of the generic operators, the processing of the working parameters and the constrained conditions of the generic algorithm are introduced. Finally, the thesis describes the embedding algorithm and the extracting algorithm in detail.
(4) In the process of the generic computed encoding, since the value of the decision variable is large compared with normal ones in this thesis, normal binary encoding is not predominant enough. Creatively, the decimal encoding method is applied in the thesis that can not only improve the operating accuracy of the generic algorithm, but reduce the calculated complexity and increase the operating efficiency.
(5) In the generic algorithm, there are three generic operators: selection operator, crossover operator and mutation operator, and there are gambling site selection, ranking selection and the best individual choice in the selection operator; single-point crossover, the two point crossover and multi-point crossover in the crossover operator; basic bit mutation, Gaussian mutation and uniform mutation in the mutation operator. In this paper, the operator of each study and experimental verification of each operator selected the best algorithms.
(6) In the generic algorithm, there are four operating parameters. As setting of the parameters and processing of the constrained conditions are associated with the accuracy, the reliability, calculating time and many other factors of the generic algorithm, and they may even influence the result quality and the system performance. Therefore, by several experiments, the thesis gives the optimal operating method.
(7) In the application of genetic algorithm, it is necessary to deal with constraints, which constitute in three types of methods: the search space limited to law, a viable solution and the penalty function method. But there has yet to find a way to deal with the general method of constraint conditions. Therefore, in terms of processing constraints, according to different application, constraints characteristics, and the computing ability we choose different approaches.
(8) Through the simulating experiments on the computer, the results show that the application of the generic algorithm can optimize the watermarking embedding and improve the graphic quality after embedding the watermarks greatly Innovative work is as follows:
(1) In the time domain, applying the generic algorithm to find optimal watermark embedding points to embed watermarks and the peak value signal to noise ratio to be the fitness function could improve the graphic quality embedded by the watermarks. Moreover, the algorithms are simple compared with previous ones, and the real-time property becomes much better.
(2) In the generic encoding, the encoding using real numbers is applied, which means taking the value representing the position of the graphic pixel points to be the decision variables. Compared with binary encoding, this method can avoid the conflicts of the embedding points caused by the crossover operation and mutation operation, increase the mathematical precision, improve the computing complexity and enhance the mathematical efficiency.
(3) The selection operators used in this thesis are to combine the Roulette Wheel Selection, the sorting selection and the optimal individual retaining, which means that to combine the proportional selection and the elitist genetic algorithm. This method not only can ensure the individuals with higher fitness value can be chosen as a priority, but can eliminate the individuals with much lower fitness value.
引文
[1]常敏,卢超,蒋明,袁春风.数字图像水印综述[J].计算机应用研究,2003(10): 1-3.
[2] I J Cox,M L Miller. The First 50 Years of Electronic Watermarking[J]. EURASIP Journal of Applied Signal Processing,2002(2): 126-132.
[3] F.A.P.Petitcolas,R.J.Anderson and M .G.Kuhn. Information Hiding[J]. A Survey. Proc. of the IEEE,special issue on protection of multimedia content,1999,87(7): 1062-1078.
[4] R. G. Schyndel,A. Z. Tirkel, and C. F. Osborne. A digital watermark Proceedings of IEEE International Conference of Image Processing[C]. Austin,Texas. 1994(2): 86-90.
[5]易开祥,石教英,孙鑫.数字水印技术研究进展[J].中国图像图形学报,2001,6 (2): 111-117.
[6]尹浩,林闯,邱锋,丁嵘.数字水印技术综述[J].计算机研究与发展,2005,42 (7): 1093-1099.
[7] G. Voyatzis,I. Pitas. The use of watermarks in the protection of digital multimedia products[J]. Proceedings of the IEEE,1999,87(7): 1197-1207,4(4): 33-46.
[8] W. Bender,D. Gruh, N. Morimoto, et al. Techniques for data hiding[J]. IBM System Journal,1996,35(2): 313-336.
[9] R. B. Wolfgang,E. J. Delp. A watermark for still images[J]. IEEE Int’l Conf. Image Processing. Lausanne,Switzerland,1996,12: 23-25.
[10] A. Z. Tirke.,C. F. Osborne,T. E. Hall. Image and watermark registration[J]. Signal Processing,1998,66(3): 373-383.
[11] R. Barnett. Digital Watermarking: Applications,Techniques and Challenges[J]. Electronics & Communication Engineering Journal,1999,11(4): 173-183.
[12] C. Walker.“Personal Communication”[J]. Muzak Corporation,2001,12(5): 3-5.
[13] K Gopalakrishnan,et al. Protocols for Watermark Verification[J]. IEEE Multimedia Magazine,2001: 66-70.
[14]刘俊,郭立.一种基于遗传算法的数字水印技术[J].微电子学与计算机,2005,22(12): 169-170.
[15]王忠,叶雄飞.遗传算法在数字水印技术中的应用[J].武汉工程大学学报,2008,30(1): 95-97.
[16]徐熙君,许曰滨.遗传算法在图像水印嵌入技术中的实现[J].青岛大学学报(自然科学版),2005,18(2): 71-74.
[17] L. Holt,B.G. Maufe and A.Wiener.“Encoded Marking of a Recording Signal”[J]. U.K. Patent,1998,5(3): 34-36.
[18] K.Tanaka,Y.Nakamura and K.Matsui. Embedding secret information Into a dithered multi-level image. IEEE Military Communication Conf[C]. USA,1990: 216-220.
[19] J M Acken. How Watermarking Adds Value to Digital Content[J]. Communications of the ACM,1998,41(7): 74-77.
[20] Xiamu Niu,Zheming Lu and Shenghe Sun. Digital image watermarking based on multiresolution decomposition[J]. IEE Electronics Letter,2000,36(13): 1108-1110.
[21] Christine I. Podilchuk,Edward J. Delp. Digital watermarking: Algorithms and applications[J]. Signal Processing Magazine,2001: 34-56.
[22] Ingemar J. Cox,J. P. Linnartz. Some general methods for tampering with watermarks[J]. IEEE Journal on Selected Areas in Communication,1998,16(4): 587-593.
[23] Ingemar J. Cox , Joe Kilian,F. Thomason Leighton,et al. Secure spread spectrum watermarking for multimedia[J]. IEEE Trans. Image Processing,1997,6(12): 1673-1687.
[24] R.B. Wolfgang and E.J. Delp. A Watermark for Digital Images. IEEE International Conference on Images Processings[C]. Lausanne,Switzerland,1996,3: 219-222.
[25] A.Z.Tirkel,C.F. Osborne and T.E. Hall. Image and Watermark Registration[J]. Signal Processing,1998,66(3): 319-335.
[26] C. I. Podilchuk,W. Zeng. Perceptual watermarking of still images[J]. IEEE Workshop Multimedia Signal Processing,Princeton NJ,1997,13: 21-24.
[27] Yu Nenghai , Cao liangliang , Fang Wen. et al. Practical analysis of watermarking capacity. Int’l Conf. Communication Technology Proceedings[C]. Beijing,2003,23: 46-48.
[28] Liu Tong , Qiu Zhengding. Attacks and evaluation in image digital watermarking[J]. Information and Control,2001,18(2): 13-15.
[29] S. Craver,N. Memon, B. L. Yeo,et al. On the invariability of invisible watermarking techniques. IEEE Int’l Conf. Image Processing[C]. Washington,CA,1997,17: 89-91.
[30] Lee SinJoo,Jung Sung-Hwan. A survey of watermarking techniques applied to multimedia[J]. IEEE Int’l Symposium Electronics,Pusan,Korea,2001,26(4): 54-55.
[31] R.G. van Schynedl,A.Z. Tirkel and C.F. Osborne. A Digital Watermark. Proceedings of IEEE International Conference on Image Processing[C]. 1994,2: 86-90.
[32] D. Kacker,T. Camis and J. P. Allebach.“Electro photographic process embedded in direct binary search”[J]. IEEE Trans. Image Processing,2002,11: 234-257.
[33]席裕庚,柴天佑.遗传算法综述[J].控制理论与应用,1996,13(6): 697-708.
[34] Holland,J.H. Concerning Efficient Adaptive Systems[J]. In Yovits,M. C.,Eds.,Self-Organizing systems,1962: 215-230.
[35] Holland,J.H. Adaptation in Natural and artificial Systems. 1975[C]. Cambridge. MA: MIT press,1992.
[36] Bremermann. H. J.. Optimization Through Evolution and Recombination in Self-Organizing Systems[J],Yovits, M. C.,Jacobi,G. T. and Goldstine,G. D.. Eds.,Spartan Books,1962: 93-106.
[37]邹燕明.遗传算法[J].名词与术语,1997(11): 57.
[38] Fogel,L. J.,Owens,A. J.,Walsh,M. J. Artificial Intelligence through Simulated Evolution[J]. New York: John Wiley,1966,10: 56-58.
[39] Fraser,A. S.. Simulation of Genetic System by Automatic Digital Computers[J].I: Introduction. Australian J. of Biol,Sci,1957,10: 484-491.
[40] Barricelli,N. A.. Sym biogenetic Evolution Processes Realized by Artificial Methods[J]. Methods,1957,9(35-36): 143-182.
[41]陈国良,王煦法,庄镇泉,王东生.遗传算法及其应用[M].北京:人民邮电出版社,1996.
[42] SrinivasM,Patnaik L M. Adaptive Probabilities of Crossover and Mutation in Genetic Algorithm[J]. IEEE Trans SystMan,Cybern,1994,24(4): 656-667.
[43] DavisL. Adaptive Operator Probabilities in Genetic Algorithms. In Proc 3rd Int Conf Genetic Algorithms[C]. 1989: 61-69.
[44] Fogarty J C. Varying the Probability of Mutation in Genetic Algorithms. In Proc 3rd Int Conf Genetic Algorithms[C]. 1989: 42-50.
[45] Whitley D,Stark weather D. GeneticⅡ: A Distributed Genetic Algorithm[J]. Expt J Theory,Art Intell,1990,2: 189-214.
[46] Michalewicz Z. Genetic Algorithm s + Date Structure = Evolution Programs[J]. Springer-Verlag,Berlin,1992,8(2): 6-9.
[47] Wright A H. Genetic algorithms for Real Parameter Optimization[J]. Foundation of Genetic Algorithms,Morgan kaufmann,Sanmateo,1991: 205-218.
[48] Brindle. A. Genetic Algorithms for Function Optimization[J]. Doctoral; Dissertation,Univ. of Alberta,1981: 25-27.
[49]陈建安,郭大伟,徐乃平,孙云芝.遗传算法理论研究综述[J].西安电子科技大学学报,1998 , 25(3): 363-367.
[50]余新宁,王文鹏,张骏.遗传算法程序的模块化设计[J].微机发展,2003,13(3): 4-6.
[51]张金奎,林荫宇,杨传忠,甘兴国,谢科.基于遗传算法的潮流多根求解方法[J].重庆大学学报(自然科学版),2000,23(1): 56-59.
[52] D. E. Goddberg. Genetic Algorithms in Search[J]. Optimization and Machine Learning Add cson- wesley Publishing Company,1989,20: 38-40.
[53] Genetic Algorithms and Their Application. Proceedings of International Conference on Genetic Algorithm[R]. 1993.
[54]唐飞,滕弘飞.一种改进的遗传算法及其在布局优化中的应用[J].软件学报,1999,10(10): 1096-1102.
[55]玄光男,程润传.遗传算法与工程设计[M].北京:科学出版社,2000.
[56]何新贵,梁久祯.目标函数梯度的遗传算法[J].软件学报,2001, 2(7): 981-985.
[57]孙艳丰,王众托.自然编码遗传算法的最优群体规模[J].信息与控制,1996,25(5): 317-320.
[58] J Craig Potts,Terri D Giddens,Surya B Yadav. The Development and Evaluation of an Improved Genetic Algorithm Based on Migration and Artificial Selection[J]. IEEE Transactions on systems. Man And Cybernetics,1994,24(1): 73-86.
[59] Muhlenbein , Schlierkamp-Voosen. Predictive models for breeder genetic algorithm: Continuous parameter optimization[J]. Evolutionary Computation,1993,1(1): 25-49.
[60]王飞朝.遗传算法编程分析[J].火控雷达技术,2005,34(2): 63-66.
[61]刘国华,包宏,李文超.用MATLAB实现遗传算法程序[J].计算机应用研究,2001(8): 80-82.
[62]杜鹏超,唐通林.数字水印研究中常用的测试指标(上)[J].电子质量,2002 (11): 7-10.
[63] S. Voloshynovskiy,S. Pereira,T. Pun, et al. Attacks on digital watermarks: Classification , estimation based attacks and bench marks[J]. IEEE Communications Magazine,2001,39(8): 118-126.
[2] I J Cox,M L Miller. The First 50 Years of Electronic Watermarking[J]. EURASIP Journal of Applied Signal Processing,2002(2): 126-132.
[3] F.A.P.Petitcolas,R.J.Anderson and M .G.Kuhn. Information Hiding[J]. A Survey. Proc. of the IEEE,special issue on protection of multimedia content,1999,87(7): 1062-1078.
[4] R. G. Schyndel,A. Z. Tirkel, and C. F. Osborne. A digital watermark Proceedings of IEEE International Conference of Image Processing[C]. Austin,Texas. 1994(2): 86-90.
[5]易开祥,石教英,孙鑫.数字水印技术研究进展[J].中国图像图形学报,2001,6 (2): 111-117.
[6]尹浩,林闯,邱锋,丁嵘.数字水印技术综述[J].计算机研究与发展,2005,42 (7): 1093-1099.
[7] G. Voyatzis,I. Pitas. The use of watermarks in the protection of digital multimedia products[J]. Proceedings of the IEEE,1999,87(7): 1197-1207,4(4): 33-46.
[8] W. Bender,D. Gruh, N. Morimoto, et al. Techniques for data hiding[J]. IBM System Journal,1996,35(2): 313-336.
[9] R. B. Wolfgang,E. J. Delp. A watermark for still images[J]. IEEE Int’l Conf. Image Processing. Lausanne,Switzerland,1996,12: 23-25.
[10] A. Z. Tirke.,C. F. Osborne,T. E. Hall. Image and watermark registration[J]. Signal Processing,1998,66(3): 373-383.
[11] R. Barnett. Digital Watermarking: Applications,Techniques and Challenges[J]. Electronics & Communication Engineering Journal,1999,11(4): 173-183.
[12] C. Walker.“Personal Communication”[J]. Muzak Corporation,2001,12(5): 3-5.
[13] K Gopalakrishnan,et al. Protocols for Watermark Verification[J]. IEEE Multimedia Magazine,2001: 66-70.
[14]刘俊,郭立.一种基于遗传算法的数字水印技术[J].微电子学与计算机,2005,22(12): 169-170.
[15]王忠,叶雄飞.遗传算法在数字水印技术中的应用[J].武汉工程大学学报,2008,30(1): 95-97.
[16]徐熙君,许曰滨.遗传算法在图像水印嵌入技术中的实现[J].青岛大学学报(自然科学版),2005,18(2): 71-74.
[17] L. Holt,B.G. Maufe and A.Wiener.“Encoded Marking of a Recording Signal”[J]. U.K. Patent,1998,5(3): 34-36.
[18] K.Tanaka,Y.Nakamura and K.Matsui. Embedding secret information Into a dithered multi-level image. IEEE Military Communication Conf[C]. USA,1990: 216-220.
[19] J M Acken. How Watermarking Adds Value to Digital Content[J]. Communications of the ACM,1998,41(7): 74-77.
[20] Xiamu Niu,Zheming Lu and Shenghe Sun. Digital image watermarking based on multiresolution decomposition[J]. IEE Electronics Letter,2000,36(13): 1108-1110.
[21] Christine I. Podilchuk,Edward J. Delp. Digital watermarking: Algorithms and applications[J]. Signal Processing Magazine,2001: 34-56.
[22] Ingemar J. Cox,J. P. Linnartz. Some general methods for tampering with watermarks[J]. IEEE Journal on Selected Areas in Communication,1998,16(4): 587-593.
[23] Ingemar J. Cox , Joe Kilian,F. Thomason Leighton,et al. Secure spread spectrum watermarking for multimedia[J]. IEEE Trans. Image Processing,1997,6(12): 1673-1687.
[24] R.B. Wolfgang and E.J. Delp. A Watermark for Digital Images. IEEE International Conference on Images Processings[C]. Lausanne,Switzerland,1996,3: 219-222.
[25] A.Z.Tirkel,C.F. Osborne and T.E. Hall. Image and Watermark Registration[J]. Signal Processing,1998,66(3): 319-335.
[26] C. I. Podilchuk,W. Zeng. Perceptual watermarking of still images[J]. IEEE Workshop Multimedia Signal Processing,Princeton NJ,1997,13: 21-24.
[27] Yu Nenghai , Cao liangliang , Fang Wen. et al. Practical analysis of watermarking capacity. Int’l Conf. Communication Technology Proceedings[C]. Beijing,2003,23: 46-48.
[28] Liu Tong , Qiu Zhengding. Attacks and evaluation in image digital watermarking[J]. Information and Control,2001,18(2): 13-15.
[29] S. Craver,N. Memon, B. L. Yeo,et al. On the invariability of invisible watermarking techniques. IEEE Int’l Conf. Image Processing[C]. Washington,CA,1997,17: 89-91.
[30] Lee SinJoo,Jung Sung-Hwan. A survey of watermarking techniques applied to multimedia[J]. IEEE Int’l Symposium Electronics,Pusan,Korea,2001,26(4): 54-55.
[31] R.G. van Schynedl,A.Z. Tirkel and C.F. Osborne. A Digital Watermark. Proceedings of IEEE International Conference on Image Processing[C]. 1994,2: 86-90.
[32] D. Kacker,T. Camis and J. P. Allebach.“Electro photographic process embedded in direct binary search”[J]. IEEE Trans. Image Processing,2002,11: 234-257.
[33]席裕庚,柴天佑.遗传算法综述[J].控制理论与应用,1996,13(6): 697-708.
[34] Holland,J.H. Concerning Efficient Adaptive Systems[J]. In Yovits,M. C.,Eds.,Self-Organizing systems,1962: 215-230.
[35] Holland,J.H. Adaptation in Natural and artificial Systems. 1975[C]. Cambridge. MA: MIT press,1992.
[36] Bremermann. H. J.. Optimization Through Evolution and Recombination in Self-Organizing Systems[J],Yovits, M. C.,Jacobi,G. T. and Goldstine,G. D.. Eds.,Spartan Books,1962: 93-106.
[37]邹燕明.遗传算法[J].名词与术语,1997(11): 57.
[38] Fogel,L. J.,Owens,A. J.,Walsh,M. J. Artificial Intelligence through Simulated Evolution[J]. New York: John Wiley,1966,10: 56-58.
[39] Fraser,A. S.. Simulation of Genetic System by Automatic Digital Computers[J].I: Introduction. Australian J. of Biol,Sci,1957,10: 484-491.
[40] Barricelli,N. A.. Sym biogenetic Evolution Processes Realized by Artificial Methods[J]. Methods,1957,9(35-36): 143-182.
[41]陈国良,王煦法,庄镇泉,王东生.遗传算法及其应用[M].北京:人民邮电出版社,1996.
[42] SrinivasM,Patnaik L M. Adaptive Probabilities of Crossover and Mutation in Genetic Algorithm[J]. IEEE Trans SystMan,Cybern,1994,24(4): 656-667.
[43] DavisL. Adaptive Operator Probabilities in Genetic Algorithms. In Proc 3rd Int Conf Genetic Algorithms[C]. 1989: 61-69.
[44] Fogarty J C. Varying the Probability of Mutation in Genetic Algorithms. In Proc 3rd Int Conf Genetic Algorithms[C]. 1989: 42-50.
[45] Whitley D,Stark weather D. GeneticⅡ: A Distributed Genetic Algorithm[J]. Expt J Theory,Art Intell,1990,2: 189-214.
[46] Michalewicz Z. Genetic Algorithm s + Date Structure = Evolution Programs[J]. Springer-Verlag,Berlin,1992,8(2): 6-9.
[47] Wright A H. Genetic algorithms for Real Parameter Optimization[J]. Foundation of Genetic Algorithms,Morgan kaufmann,Sanmateo,1991: 205-218.
[48] Brindle. A. Genetic Algorithms for Function Optimization[J]. Doctoral; Dissertation,Univ. of Alberta,1981: 25-27.
[49]陈建安,郭大伟,徐乃平,孙云芝.遗传算法理论研究综述[J].西安电子科技大学学报,1998 , 25(3): 363-367.
[50]余新宁,王文鹏,张骏.遗传算法程序的模块化设计[J].微机发展,2003,13(3): 4-6.
[51]张金奎,林荫宇,杨传忠,甘兴国,谢科.基于遗传算法的潮流多根求解方法[J].重庆大学学报(自然科学版),2000,23(1): 56-59.
[52] D. E. Goddberg. Genetic Algorithms in Search[J]. Optimization and Machine Learning Add cson- wesley Publishing Company,1989,20: 38-40.
[53] Genetic Algorithms and Their Application. Proceedings of International Conference on Genetic Algorithm[R]. 1993.
[54]唐飞,滕弘飞.一种改进的遗传算法及其在布局优化中的应用[J].软件学报,1999,10(10): 1096-1102.
[55]玄光男,程润传.遗传算法与工程设计[M].北京:科学出版社,2000.
[56]何新贵,梁久祯.目标函数梯度的遗传算法[J].软件学报,2001, 2(7): 981-985.
[57]孙艳丰,王众托.自然编码遗传算法的最优群体规模[J].信息与控制,1996,25(5): 317-320.
[58] J Craig Potts,Terri D Giddens,Surya B Yadav. The Development and Evaluation of an Improved Genetic Algorithm Based on Migration and Artificial Selection[J]. IEEE Transactions on systems. Man And Cybernetics,1994,24(1): 73-86.
[59] Muhlenbein , Schlierkamp-Voosen. Predictive models for breeder genetic algorithm: Continuous parameter optimization[J]. Evolutionary Computation,1993,1(1): 25-49.
[60]王飞朝.遗传算法编程分析[J].火控雷达技术,2005,34(2): 63-66.
[61]刘国华,包宏,李文超.用MATLAB实现遗传算法程序[J].计算机应用研究,2001(8): 80-82.
[62]杜鹏超,唐通林.数字水印研究中常用的测试指标(上)[J].电子质量,2002 (11): 7-10.
[63] S. Voloshynovskiy,S. Pereira,T. Pun, et al. Attacks on digital watermarks: Classification , estimation based attacks and bench marks[J]. IEEE Communications Magazine,2001,39(8): 118-126.