摘要
20世纪末,数字多媒体的安全问题成为信息安全领域的研究热点。文本图像作为一类特殊的数字多媒体,在社会生活中具有举足轻重的地位。在当前脆弱的互联网安全机制下,某些重要文本图像的使用有很多安全隐患。因此,有必要对文本图像的认证技术进行深入研究,用以保障这些重要文本的安全使用。
本文首先全面介绍并分析了当前文本图像认证领域的研究现状,然后分别对二值文本图像认证和鲁棒文本图像认证进行了深入研究,主要成果如下：
(1)针对当前二值文本图像认证算法无法有效解决图像均匀区篡改认证问题,提出一种基于短数字签名的二值文本图像认证算法。该算法通过对图像的均匀区分布情况提取长度很短的数字签名,有效的解决了图像均匀区的篡改认证问题；并且,算法在保证良好不可见性的前提下,对非均匀区有75%的理论篡改检测概率。
(2)经过观察和分析发现,二值文本图像16模块中心像素点翻转后并不会改变图像行、列扫描向量的游程个数。基于这一特性,设计了一种新的基于数字水印的二值文本图像认证算法,实现了二值文本图像均匀区的认证保护和较高精度的篡改定位。实验表明,算法能有效认证图像均匀区,并具有较高篡改检测率和良好不可见性。
(3)针对汉字的骨架结构特性,设计了一种文本图像特征提取的新方法。实验表明,这种提取文本图像特征的方法具有一定的抗噪能力,可用于文本图像的内容认证和篡改定位。
(4)改进了一种小波域水印嵌入方法,通过减少修改参数的个数来改善原算法不可见性较差这一缺点。实验表明,改进后嵌入算法的鲁棒性较原算法略有下降,但不可见性远优于原算法,可用于文本图像的版权保护和内容认证。
(5)首次分析了对灰度文本图像进行内容认证的必要性,并结合(3)、(4)的结论,提出一种鲁棒文本图像内容认证方案。实验表明,该方案对灰度文本图像有50%的篡改检测概率、并具有较好不可见性和一定的鲁棒性。
In the late 20th century, security issues on digital multimedia become a hot research in the information security field. As a special kind of digital media, text image plays a decisive role in social life. In the current fragile internet security mechanism, there are lots of security problems in using some important text images. Therefore, it is necessary to study text image authentication in-depth in order to protect these important texts.
Firstly, the current research status about text image authentication is introduced and analyzed in the first two chapters of this paper. Then binary text images authentication and robust image authentication is studied in-depth respectively in the last three chapters. The main results to the research are as follows:
(1) Because current binary text image authentication algorithm cannot effectively solve the problem that tampering authentication to the uniformity of image, a new binary text image authentication algorithm based on short digital signature is proposed. In this new algorithm, the distribution of the uniform area in the image is extracted to a short signature, which can effectively solve the issue tampering authentication to the uniformity of image. With good invisibility, the algorithm has the theoretical probability of tamper detection 75% to the non-uniform area of the image.
(2) Through observing and analyzing, it is found that turning the centre pixel of The-16-Blocks does not change the number of runs to the row or column scan vectors of the binary text image. Based on this feature, a new binary text image authentication algorithm based on watermark is designed, by which the uniform area of the image can be certified and high accuracy positioning can be achieved. The experiments show that the algorithm can effectively certificate the uniform areas of the image, with a higher tampering detection rate and good invisibility.
(3) Considering the characteristics of the character structure, a new text image feature extracting method is designed. Experimental results show that this image feature extracting method has a certain anti-noise ability and it can be used for text image content authenticating and tamper locating.
(4) A wavelet domain watermark embedding method is improved. The poor invisibility of the original algorithm is highly improved by reducing the number of parameters to modify. Experimental results show that the robustness of the improved embedding method is slightly worse than the original one, but the invisibility is much better than the original one. The improved algorithm can be used for protecting the copyright and authenticating the content of the text image.
(5) The necessity of authenticating the content of the gray text image is proposed in the paper. Combing with the conclusions of (3) and (4), a robust text image content authentication algorithm is proposed. Experiments show that the algorithm has a tampering detecting probability 50% to the gray text image, and the invisibility and robustness works well.
引文
[1]孙圣和,陆哲明,牛夏牧.数字水印技术及应用[M].北京：科学出版社,2004：1-5
[2]李庆诚,李缓挥.网络出版中版权保护技术—文本数字水印的研究[J],计算机工程与应用,2004,8：136-156
[3]王慧琴,李人厚.二值文本数字水印技术的研究与仿真[J],系统仿真学报,2004，16(3)：521-524
[4]张黎.基于小波变换的二值图像技术研究[D],山东：山东师范大学,2007
[5]D. Huang, H. Yan. Interword Distance Changes Representedby SineWaves for Watermarking Text Images [J]. IEEE Trans. On Circuits and Systems for Video Technology,2001,11(12):1237-1245
[6]Kim H Y, Afif A. Secure authentication watermarking for binary images[C].In Proc.Sibgrapi-Brazilian Symp.on Comp. Graph.and Image Proc.2003:199-206
[7]Young-Won Kim, Kyung-Ae Moon, and Il-Seok Oh. A Text Watermarking Algorithm based on Word Classification and Inter-word Space Statistics [J]. IEEE Proceedings of the Seventh International Conference on Document Analysis and Recognition,2003
[8]A.T.S. Ho, N. B. Puhan, P. Marziliano, A. Makur, Y. L. Guan. Perception Based Binary Image Watermarking[C]. IEEE International Symposium on Circuits and Systems (ISCAS), Vancouver, Canada, May (2004):23-26
[9]Awan, I, Gilani, S.A.M, Shah, S.A, Utilization of Maximum Data Hiding Capacity in Object-Based Text Document Authentication Intelligent Information Hiding and Multimedia Signal Processing [J]. IIH-MSP'06. International Conference on Dec. 2006:597-600
[10]Ru yu Liu, et al. An Image Authentication Scheme Based on Sliding Window[C]. 2008 Chinese Control and Decision Conference (CCDC2008):2937-2940
[11]M. Wu, E. Tang, and B. Liu. Data hiding in digital binary image [J]. in Proc. IEEE Int. Conf. Multimedia and Expo., New York,2000:393-396.
[12]M. Wu and B. Liu. Data hiding in binary images for authentication and annotation [J]. IEEE Trans. Multimedia, Vol.6, No.4, Aug.2004:528-538
[13]H.Lu, A.C.Kot and J.Cheng. Secure data hiding in binary document images for authentication [J]. Proc.ISCAS'04, Vol.3,2004:806-809
[14]Chao Yu, Xudong Zhang, Watermark embedding in binary images for authentication Signal Processing [J].2004 Proceedings. ICSP'04.2004 7th International Conference on Vol.1,31 Aug.-4 Sept.2004:865-868
[15]Huijuan Yang, Kot A.C. Data Hiding For Binary Images Authentication By Considering A Larger Neighborhood [J]. IEEE International Symposium on Circuits and Systems,27-30 May 2007:1269-1272
[16]Huijuan Yang, Kot A.C. Date hiding for text document image authentication by connectivity-preserving [J]. in Proc. IEEE Int. Conf. on Acoustics, Speech, and Signal Processing, Philadelphia, PA, Mar.18-23,2005, vol.2:505-508
[17]Huijuan Yang, Kot A.C. Two-Layer Binary Image Authentication With Tampering Localization [J]. IEEE International Conference on Acoustics, Speech and Signal Processing,2006, Volume 2, May 2006:14-19
[18]Puhan, N.B., Ho, A.T.S. Binary document image watermarking for secure authentication using perceptual modeling Signal Processing and Information Technology[C].2005 Proceedings of the Fifth IEEE International Symposium on18-21 Dec.2005:393-398
[19]Liu Hong mei, Rui Wei, Huang, Jiwu. Binary Image Authentication using Zernike Moments Image Processing [J]. ICIP 2007. IEEE International Conference on Volume 1, Sept.16 2007-Oct.19,2007:385-389
[20]C.K.Hsiang, C.Y.Yu and T.C.Yu. A Secure Data Hiding Scheme for Two-color Images[C].the 5th IEEE Symposium on Computers and Communications (ISCC'00), July 3-6,2000:750-755
[21]李赵红,侯建军.基于等级结构的二值文本图像认证水印算法[J].自动化学报,2008,34(8)：841-848.
[22]张小华.基于数字水印的图像认证技术研究[D].西安：西安电子科技大学,2004
[23]Sheng Huang, Jian Kang Wu. Optical Watermarking for Printed Document Authentication [J]. IEEE Transaction Information for Forensics and Security, vol.2, No.2, June 2007:164-173.
[24]谭论正,朱宁波,孙星明.基于文档图像的抗打印—扫描数字水印技术[J].计算机应用研究,Dec 2007,Vol 24,No.12：200-203
[25]宋玉杰,刘瑞祯,谭铁牛等.数字水印技术在印刷品防伪中的应用[J].中国图象图形学报,2001,6(5)：450-454
[26]Patrick Bas, Jean-Marc Chassery, Benoit Macq. Geometrically Invariant Watermarking Using Fearture Points. IEEE Trans. Image Processing, vol.11 No.9,Sept.2002:1014-1028
[27]Xiangui Kang, Jiwu Huang, Yun Q. Shi. an Image Watermarking Algorithm Robust to Geometric Distortion[J]. Proc. Of IWDW 2002, vol.2613,2002:212-213
[28]牛少彰,伍宏涛,谢正程等.抗打印扫描数字水印算法的鲁棒性[J].中山大学学报(自然科学版),2004,43：1-4
[29]李刚,杨杰.一种基于二值印刷图像的数字水印方案[J].上海交通大学学报,2005,39(4)：570-573
[30]陈萍.鲁棒的印刷图像水印研究[D].北京：北京交通大学,2006
[31]Jay Fraser, President Tracer Detection Technology Corp, Exploiting Random Patterns of Optically Readable Materials to Ensure Authentication of Media & Substrates Documents[R],2008, from IEEE Explore. Restrictions.
[32]L. Haiping, W. Jian, A. C. Kot and Y. Q. Shi. An Objective Distortion Measure for Binary Document Images Based on Human Visual Perception [J].16th International Conference on Pattern Recognition,2002,4:293-242.
[33]Y. C. Tseng and H.-K. Pan. Data hiding in 2-color images [J]. IEEE Trans. Comput, vol.51, no.7, Jul.2002:873-878
[34]朱从旭,陈志刚.一种灵敏的文本图像认证混沌脆弱水印技术[J].小型微型计算机系统,27(1),2006：151-154
[35]H. Yang and A. C. Kot. Data hiding for bi-level documents using smoothing techniques [J]. in Proc. IEEE Int. Symp. Circuits Systems, ISCAS'04, May 2004, vol.5:692-695
[36]Hui juan Yang and Alex C.Kot. Pattern-Based Data Hiding for Binary Image Authentication by Connectivity-Preserving [J]. Proc. IEEE Transactions on Multimedia, Vol.9, No.3, Apr 2007:475-485
[37]李晓.基于汉字结构的中文文本水印技术研究[D].西安：西北大学,2009。
[38]张静,张春田.二值图像数字水印技术综述[J].计算机工程,Vol 31, No.3, February 2005:151-158
[39]温晓明.二值文本图像数字水印研究[D].山东：山东大学,2007
[40]孟妍妍.二值图像隐写分析方法研究[D].西安：西北大学,2009
[41]冀芊茜.抗打印扫描数字水印算法研究及其在证件防伪中的应用[D].长春：吉林大学,2007
[42]商艳红,郑铮.基于数字水印技术的商业票据防伪新方法[J].图像处理,文章编号1008-0570(2007)01-3-0281-03,2007
[43]李俊兰,周晶.一种对印刷图像进行版权保护的算法[J].计算机工程与应用,2004增刊
[44]孙妮娜,李黎.一种用于证件防伪的数字水印算法[J].杭州电子科技大学学报,Vol.27, No.6, Dec.2007:83-88
[45]郭森,柳伟.基于粗糙理论的票据打印字符识别系统[J].计算机与数字工程,Vol 36, No.8,2008(8):70-73
[46]http://baike.baidu.com/view/1624289.html
[47]黄美茹.数字水印技术在版权保护和文本认证中的应用研究[D].西安：西北大学,2008
[48]Wei Feng, Zhi-Qiang Liu. Region-Level Image Authentication Using Bayesian Structural Content Abstraction [J]. IEEE Transactions on Image Processing, Vol.17, No.12, Dec 2008:2413-2424
[49]Tie gang Gao, Qiao lun Gu. A Novel Watermarking-Based Reversible Image Authentication Scheme[C]. International Conference on Intelligent Information Hiding and Multimedia Signal Processing,2008:1269-1272
[50]Li Chuan mu, Liu Lili. An Image Authentication Scheme with Localization and Recovery[C].2008 Congress on Image and Signal Processing:669-672
