一種新的基于中國(guó)剩余數(shù)定理的多媒體內(nèi)容認(rèn)證水印算法（英文）

1、Digital Signal Processing 20 (2010) 442–453Contents lists available at ScienceDirectDigital Signal Processingwww.elsevier.com/locate/dspA novel CRT-based watermarking technique for authentication of multimedia contentsJa

2、gdish C. Patra a,?, A. Karthik a, Cedric Bornand ba School of Computer Engineering, Nanyang Technological University, Nanyang Avenue, Singapore b HEIG-VD, University of Applied Sciences, Yverdon-les-Bains, Switzerlanda r

3、 t i c l e i n f o a b s t r a c tArticle history:Available online 28 July 2009Keywords:Digital watermarkingChinese remainder theoremSingular value decompositionDigital watermarking techniques have been proposed as a sol

4、ution to the problem ofcopyright protection of multimedia data. In this paper, we propose a novel Chineseremainder theorem (CRT)-based technique for digital watermarking. The use of CRTfor this purpose provides additiona

5、l security along with resistance to some familiarattacks. We have shown that this technique is quite resilient to addition of the noise.We have compared performance of the proposed technique with recently reported twosin

6、gular value decomposition (SVD)-based watermarking techniques and shown its superiorperformance in terms of tampering assessment function (TAF), computational efficiencyand peak signal to noise ratio (PSNR). For example,

7、 the embedding time of the proposedCRT-based scheme is 6 and 3 times faster than the SVD-based Schemes 1 and 2,respectively. This technique can also be applied to document, audio and video contents. © 2009 Elsevier

8、Inc. All rights reserved.1. IntroductionToday’ s information driven economy is dominated by the tremendous growth of the Internet and explosion of day-to-daydata processing with huge amount of multimedia data. Easy avail

9、ability of content-editing software, mobile and compactdigital devices and the Internet, make the digital lifestyle of common man quite different from that of few years ago. Digitalmultimedia contents, e.g., text, image,

10、 video and audio, can be easily altered, stored or transmitted to any point of the globeinstantly. However, multimedia digital content owners are skeptical of putting their content on the Internet due to lack ofintellect

11、ual property protection available to them. In order to address this situation, digital watermarking is indeed one ofthe solutions to protect the ownership of these contents. In digital watermarking techniques, some digit

12、al signature thatis unique to the owner or copyright information is embedded into the host multimedia content. The signature embeddedremains invisible and imperceptible and cannot be removed easily even under certain man

13、ipulations, e.g., addition of noise,compression, tampering and scaling operations. Only the authorized recipient of the digital content can extract the water-mark from the watermarked content with the knowledge of some k

14、ey information. In this way security, content integrityand intellectual protection can be provided to the owner. In this direction, starting from mid-1990, several researchers havereported many watermarking techniques in

15、 spatial and transform domains [1–3].Some important properties of an effective watermarking scheme are [1–3]: (i) imperceptibility: there should not be anynoticeable difference between a watermarked content and the origi

16、nal, (ii) robustness: the embedded watermark should beable to withstand to some extent of content manipulation. Any attempt to destroy a watermark should also void the wa-termarked content and (iii) trustworthiness: no o

17、ther watermark other than the embedded watermark should be extracted* Corresponding author.E-mail addresses: aspatra@ntu.edu.sg (J.C. Patra), AL0001IK@ntu.edu.sg (A. Karthik), cedric.bornand@heig-vd.ch (C. Bornand).1051-

18、2004/$ – see front matter © 2009 Elsevier Inc. All rights reserved.doi:10.1016/j.dsp.2009.07.004444 J.C. Patra et al. / Digital Signal Processing 20 (2010) 442–453Fig. 1. Watermarking Scheme 1. The change in rank re

19、sults in a shifted Panda in the extracted watermark: (a) original 16 × 16 watermark, (b) extracted16 × 16 watermark, (c) difference between original and extracted watermarks.for embedding watermark bits. This r

20、eliance causes the scheme to be less resistant to attacks, which alters the rank of theimage blocks. The host image is a grey-scale image and the watermark is a binary image. First, the host image is dividedinto blocks o

21、f n × n pixels. A single watermark bit is embedded in a block. The blocks are chosen randomly using a pseudo-random number generator (PRNG) based on their rank. The blocks with higher ranks are selected first before

22、 those of lowerranks.In order to embed a watermark bit in a block the following procedure is adopted. We first perform a SVD transformationon the selected block. Let c1 and c2 denote the second and the third elements (u2

23、1 and u31) of the first column of U matrix,respectively. The embedding rules are as follows:To embed a watermark bit ‘ 1,’ the value of c1 ? c2 should be positive and its magnitude is greater than a strengthfactor, s. If

24、 this condition is not satisfied, c1 and c2 are modified as c? 1 and c? 2, respectively, and are given byc? 1 = a + s/2,c? 2 = a ? s/2, (2)where a = (|c1|+ |c2|)/2.To embed a watermark bit ‘ 0,’ the value of c1 ? c2 shou

25、ld be negative and its magnitude is greater than the strengthfactor, s. If this condition is not satisfied, c1 and c2 are respectively modified as c? 1 and c? 2 given byc? 1 = a ? s/2,c? 2 = a + s/2. (3)To reconstruct th

26、e watermarked block, an inverse SVD transformation is performed on the modified U matrix withthe original D and V matrices. This watermarked block then replaces the original selected block in the host image. Thisembeddin

27、g process is repeated until all the watermark bits have been embedded.In this scheme, the selection of image blocks based on higher ranks would lower the distortion level of the watermarkedimage. However, the rank is not

28、 a reliable feature as it is not stable. The rank of the selected block could change after mod-ifications are made to the elements c1 and c2. Therefore, without any tampering to the watermarked image, the extractedwaterm

29、ark could be corrupted due to a change in rank of the block. Fig. 1 shows the corruption of the extracted watermarkdue to changes in rank. In general, when the strength factor increases, the likelihood of changing of the

30、 rank of a blockincreases. This leads to a higher level of corruption and distortion in the watermark and watermarked image, respectively.The initial steps of the extraction procedure are similar to that of the embedding

31、 process up to and including theselection of the coefficients c1 and c2. The value of c1 ? c2 determines the value of the extracted watermark bit. A positivedifference indicates that the watermark bit is a ‘ 1.’Whereas,

32、a negative difference would imply that a watermark bit ‘ 0’isextracted.2.2. Watermarking Scheme 2The embedding procedure of the scheme proposed by Patra et al. [15] is an improved version of Scheme 1. The hostimage is di

33、vided into a number of super-blocks that is equal to the number of watermark bits. From each super-block,a sub-block is selected which is transformed via SVD to embed the watermark. Instead of embedding exclusively in th

34、eU matrix, the V matrix is also utilized. A counter is introduced to allow random selection of embedding the watermark bitin either U or V matrix to improve the security and robustness.In addition, the coefficients (c1 a

35、nd c2) are selected randomly from the first column of the U or V matrices using a PRNG.This flexibility gives an edge over the scheme proposed by Chang et al. [14], as malicious tampering cannot be targeted atthe two mod

36、ified coefficients due to their random selection. However, to minimize the distortion to the watermarked image,the position of c2 must be consecutive to c1. Experiments show that the watermarked image quality reduces as

37、the distancebetween the modified coefficients is increased. The modifications of c1 and c2 are similar to the procedure in Scheme 1. Thereconstruction of the new block, re-placement of the block into the image, and extra

38、ction procedure are also similar to thatof Scheme 1.This scheme removes the reliance on rank and also improves security by using random selection of coefficients forembedding the watermark bits. The weakness of this sche