M. Kutter and F. Petitcolas, "A fair benchmark for image watermarking systems," in Security and Watermarking of Multimedia Contents, ser. Proceedings of SPIE, vol. 3657, January 1999, pp. 226--239.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....of a legally obtained image, or hostile, such as an attempt by a multimedia pirate to destroy a watermark before re selling watermarked data. Examples of attacks include compression, linear filtering, geometric transformations, and D A A D conversion. Some extensive lists appear in [3] and [7], but it is impossible to name all of the potential attacks. Instead, 8] provides a set of conceptual attack categories. In the present paper, we take a theoretical approach and only consider attacks that attempt to remove a watermark or to confuse the watermark detector by linear, ....

....a watermark remains. Part of the difficulty in answering the question is that robustness is easy to postulate but hard to measure. Currently, it is still difficult to quantify the detectability of an attacked watermark and the quality of the attacked data. Some initial attempts have been made in [7] and [12] They are based on selecting a distortion measure, performing a battery of attacks on different watermarks, and measuring quantities such as the probability of error after each attack. They propose a methodology for evaluating robustness experimentally, but they are specific to a given ....

M. Kutter and F. A. P. Petitcolas, "A fair benchmark for image watermarking systems," in Proc. SPIE, Security &Watermarking Multimedia Contents, vol. 3657, San Jose, CA, Jan. 1999, pp. 226--239.

....image. Many schemes, however, set ipeak to the value of 255. For an 8 bit image, i.e. an image in which each pixel value is represented by 8 bits, 255 is the largest value the function i(x, y) can take. A peak signal to noise ratio of 38 dB and larger reflects an acceptable image qual 16 ity [14]. In the implemented system the relationship between the embedding strength and the resulting PSNR has been examined and the parameter has been chosen so that watermark invisibility is ensured. 3.2. Generation of the Watermark Sequence The watermark information consists of a sequence of ....

....the capacity of the watermarking channel. The peak signal to noise ratio as defined in equation 3.6 is a measure of the degradation in the original image introduced by the watermark as well as by other contaminations. An acceptable image quality is designated by a PSNR value of 38 dB and larger [14]. The watermark to document ratio WDR on the other hand is the signal to noise ratio from the perspective of the embedded watermark signal: the watermark is the desired signal while the host image acts as noise. The WDR is defined as WDR(dB) 10 log (5.1) where i(x, y) and iw(X, y) are the ....

Kutter M. and Petitcolas F. A. P. "A fair benchmark for image watermarking systems," Proc. of Electronic Imaging, Vol. 3657, pp. 226 239.

....linear dependency can be observed. actual watermarking method used to illustrate it. Deployment in any other watermarking method is basically identical. The theoretical grounding of LSB space key search is not trivial in the general case, as related research by Szpankowski [15] and others [16], suggests, and is also outside the scope of this paper. Nevertheless we introduce a simple experiment as a proof of concept. Experiment. Given the class of LSB marking algorithms [1] a certain predefined desired watermark w and digital images with LSB spaces (e.g. of size n = 1024 bits, an ....

M. Kutter and F. A. P. Petitcolas. A fair benchmark for image watermarking systems. In citeseer. nj.nec.com/article/kutter99fair.html, pages 226-- 239.

....Defining a unified watermarking evaluation metric is a non trivial task. Most domain specific metrics are derived from the concept of watermarking capacity and do not directly relate to the main purpose of watermarking per se, i.e. claiming ownership in court. Theoretical approaches [3] [10] [11] 12] 16] explore the broader area of steganography and information hiding in a generic manner. Proof of ownership is usually achievable by demonstrating that the particular piece of data exhibits a certain rare property (read hidden message or watermark ) usually known only to Alice ....

Martin Kutter and Fabien A. P. Petitcolas. A fair benchmark for image watermarking systems. In citeseer.nj.nec.com/article/kutter99fair.html, pages 226--239.

....unacceptable embedding distortion and highpass watermarks would be susceptible to attack; as a compromise these authors advocated the used of bandpass watermarks. Most of the early (and current) work in watermarking has been applied, with robustness and imperceptibility evaluated experimentally [32]. Many attacks consist of additive noise, compression (e.g. MP3 for audio, JPEG for images, and MPEG 2 for video) or geometric transformations such as rotation, shifting, and scaling [35] Recently, more theoretical approaches have attempted to provide watermarking, and the larger field of ....

....attack for a given watermark power spectrum ### ####. Under the assumptions of IID RVs, a Gaussian original, and MSE distortion, it was shown in [34] that the optimum attack among all possible attacks consists of scaling and additive Gaussian noise. The attack model (2) thus extends the attack in [32] by adding memory. Hence, for x#### Gaussian and MSE distortion, the attack we derive will be optimum among all attacks. The attacker s problem is to find ##### and # ## #### to minimize # # ## subject to # # # # (Problem 1) This problem can be solved by the calculus of variations; the details ....

M. Kutter and F. A. P. Petitcolas, "A fair benchmark for image watermarking systems," in [4], pp. 226-- 239.

....images. In this paper, a new watermarking method is presented exploiting both the HVS and the statistical nature of an image, in order to watermark it. The robustness of the method was tested with over 250 images under several attacks. The method is evaluated using the model of KutterPetitcolas [7]. PSNR was used for the measurement of the difference between the original and the watermarked image. This paper is structured as follows: Section 2 gives an overview of the proposed method. Section 3 presents a set of experiments and their results. Finally, Section 4 discusses the results and ....

....A watermark, as discussed in Section 2, modifies the original image. The main goal of any watermarking method is to minimize the modifications caused to the original image, so as not to be detected by the observer. In order to test the way in which this method affects the original image [7], we used three types of tests: a) measurements of Peak Signal to Noise Ratio (PSNR) between the original and the watermarked image, b) measurements of similarities in each one of the 24 m x n bit matrices and (c) tests using the HVS. 0 20 40 60 80 100 120 140 Our Method Random Figure ....

M. Kutter, F.A.P. Petitcolas, A fair Benchmark for Image Watermarking Systems, Proceedings of SPIE: Security and Watermarking of Multimedia Contents,Volume 3657, pp. 226-239, San Jose, California, January, 1999.

....watermarking audio, image, and video, and comprehensive surveys of these technologies may be found in [2] and [3] However, the literature lacks an effective means of comparing the different approaches. An evaluation framework was recently described, but is limited to digital image watermarking [4]. The goal of this paper is to present an algorithmindependent set of criteria for quantitatively comparing the performance of digital watermarking algorithms. This framework is then used to evaluate a selection of five audio watermarking algorithms from the literature. The paper is organized as ....

M. Kutter and F. Petitcolas, "Fair benchmark for image watermarking systems," Proceedings of SPIE Security and Watermarking of Multimedia Contents, vol. 3657, pp 226 239, 1999.

....acceptable embedding distortion and highpass watermarks would be susceptible to attack; as a compromise these authors advocated the used of bandpass watermarks. Most of the early (and current) work in watermarking has been applied, with robustness and impercep tibility evaluated experimentally [32]. Many attacks consist of additive noise, compression (e.g. MP3 for audio, JPEG for images, and MPEG 2 for video) or geometric transformations such as rotation, shifting, and scaling [35] Recently, more theoretical approaches have attempted to provide watermarking, and the larger field of ....

....attack for a given watermark power spectrum ww ( Under the assumptions of IID RVs, a Gaussian original, and MSE distortion, it was shown in [34] that the optimum attack among all possible attacks consists of scaling and additive Gaussian noise. The attack model (2) thus extends the attack in [32] by adding memory. Hence, for x[ff] Gaussian and MSE distortion, the attack we derive will be optimum among all attacks. The attacker s problem is to find G( and vv( to minimize Dx subject to C = Ct (Problem 1) This problem can be solved by the calculus of variations; the details appear in ....

M. Kutter and F. A. P. Petitcolas, "A fair benchmark for image watermarking systems," in [4], pp. 226- 239.

....with almost all current watermarking technologies is that they fail to recover a watermark from random bending geometrical distortions, known as the random bending attack (RBA) The RBA was first introduced by F. Petitcolas in the benchmarking tool Stirmark to model printing scanning artifacts [1]. If today watermarking technologies resist in practice against printing scanning, unfortunately however the RBA attack still remains an essential problem for almost all existing watermarking methods. The practical danger of this attack consists in the fact that the attacker can apply it against ....

....macro block, approximated as an affine transform mn A (up) The same, considering the reference watermark (the black squares) which can be used to locally resynchronize the watermark information (down) 5. RESULTS We tested our approach based on the Stirmark 3.1 benchmark of F. Petitcolas [1] using 6 standard images. The results are shown in Table 1, by marks between 0 (no watermark decoded) to 1 (watermark resisted all attacks) The watermark was decoded from all randomly distorted images, significantly increasing the total score up to 0.996 over 1. Today, no known algorithm presents ....

M. Kutter and F.A.P. Petitcolas, "A fair benchmark for image watermarking systems", Proceedings of SPIE: Security and Watermarking of Multimedia Content, vol. 3657, pp. 219-239, San Jose, CA, USA, January 1999.

....wPSNR=27.9dB, TPE=7.87, NB1=128, NB2=0, d) watermarked was added to the cover image without masking: PSNR=24.61dB, wPSNR=29.3dB, TPE=9.27, NB1=146, NB2=3 generation benchmark. We note that the benchmark we propose is not intended to replace the benchmark proposed by Kutter and Petitcolas [36], but rather to complement it. While their benchmark heavily weights geometric transformations and contains non adaptive attacks, the benchmark we propose includes models of the image and watermark in order to produce more effective attacks. 12.1 A New Benchmark The benchmark consists of six ....

M. Kutter and F. A. P. Petitcolas. A fair benchmark for image watermarking systems. In Electronic Imaging '99, Security and Watermarking of Multimedia Contents, volume 3657, pages 219--239, San Jose, CA, USA, January 1999.

....oblivious. For strictly oblivious watermarking, the watermark detector may not even know the size of the original. The received image may have been resized, rotated, cropped, undergone some histogram modification, and probably some geometric distortion such as those introduced by StirMark [8]. If by some way, a watermarking scheme is able to detect the watermark in a strictly oblivious fashion (in the face of all possible attacks) the watermarking scheme would have drastically reduced degrees of freedom for the choice of the signature. Reduced degrees of freedom implies lower ....

M. Kutter, F.A.P. Petitcolas. "A Fair Benchmark for Image Watermarking Systems", Electronic Imaging, vol 3657, pp. 226-239, San Jose, CA, USA, January 1999.

....of Technology New Jersey Center for Multimedia Research University Heights, Newark, NJ, 07102. ABSTRACT The goal of this study is to investigate the performance of a previously proposed watermarking scheme [1] for watermarked images distorted by di erent geometrical attacks such as StirMark [2] and swirl. Although the geometrical distortions are not visually noticeable, the watermark cannot be detected due to the high mean square error between the original and attacked images. In order to extract the watermark from distorted images, the e ects of these attacks must be minimized. This ....

....d Figure 1: Watermark Embedding and Detection Protocol tacks, and possibly undo them in order to extract the watermark with a reasonably high degree of certainty. Ref. 1] identi ed pixel scaling, histogram modi cation, and imperceptible geometric distortion (such as those introduced by StirMark [2]) as the principal ways in which such fake originals can be created. The watermark detection algorithm therefore needs regulated algorithms for undoing or minimizing the e ects of such attacks. The block diagram of the proposed protocol is shown in Figure 1. The overall protocol consists of 1. A ....

M. Kutter and F. A. P. Petitcolas. \A Fair Benchmark for Image Watermarking Systems", Electronic Imaging: Security and Watermarking of Multimedia Contents, vol. 3657, pp. 226-239, San Jose, CA, USA, January 1999.

....data is not able to survive those compression methods, the content loses its value. Therefore, more useful data hiding techniques should utilize holes which are very dicult to plug. Figure 3 depicts the original 256 256 Goldhill image, its histogram reshaped version, and image after StirMark [2] (StirMark is a watermark attack software that introduces imperceptible geometric distortions in the image) Though the second and third images are very close to the original in visual delity, their PSNRs are 20 and 19 dB respectively It is clear that signi cant amounts of distortion (in the MSE ....

M. Kutter and F. A. P. Petitcolas. \A Fair Benchmark for Image Watermarking Systems",Proceedings of Electronic Imaging '99, Security and Watermarking of Multimedia Contents, San Jose, California, vol. 3657, pp. 226-239, January 1999.

....image degradation also increases as a function of the embedding strength. This means that the watermarking system designer will face a trade off between robustness and visual quality. In order to evaluate visual distortions introduced by the insertion process, two pixel based metrics were used [17]: i. Signal to Noise Ratio (SNR) ii. Peak Signal to Noise Ratio (PSNR) The SNR is expressed as ( # D D D q7 TIS , 2 , 2 10 ) log 10 ) 6.2) The PSNR is given by ( # D D D I q7 QTIS , 2 2 2 10 ) ....

....PSNR curves related to the experiment. Again, SRN and PSNR curves monotonically decrease when the perceptual parameter is increased. This was expected, since changes introduced in each pixel value increase with a. 6. 4 Attacks Results obtained with the proposed method for the most common attacks [17] on watermarks are presented in this section. Experiments were conducted for the following attacks: i. Translation 55 ii. Rotation iii. Scaling iv. JPEG compression v. Noise addition vi. Low pass filtering: median and average filters vii. Histogram equalization 6.4.1 Translation Recall ....

M.Kutter and F.A.P. Petitcolas, "A fair benchmark for image watermarking systems", Electronic Imaging '99 -- Security and Watermarking of Multimedia Contents, vol. 3657, January 1999, San Jose, United States.

....to exploit human perception, e.g. the human auditory system in case of audio watermarks and the human visual system (HVS) in case of image and video watermarks. Third, we discuss the current status of image watermarking benchmarks. We briefly present Fabien Petitcolas Stirmark benchmarking tool [1]. Next, we consider the benchmark proposed by the University of Geneva Vision Group that contains more deliberate attacks. Finally, we summerize the current work of the European Certimark project, whose goal is to accelerate efforts from a number of research groups and companies in order to ....

....information. The detector could recover the embedded watermark information when perfect synchronization is regained. However, the complexity of the required synchronization process might be too great to be practical. For image watermarking, the most known benchmarking tools, Unzign and Stirmark [1], integrate a variety of geometric attacks. Unzign introduces local pixel jittering and is very efficient in attacking spatial domain watermarking schemes. Stirmark introduces both global and local geometric distortions. We give a few more details about these attacks later in this paper. However, ....

[Article contains additional citation context not shown here]

M. Kutter and F. Petitcolas, "A fair benchmark for image watermarking systems," Electronic Imaging 199: Security and Watermarking of Multimedia Content,Vol.3657 of SPIE Proceedings, San Jose, California USA, 25-27 January 1999.

....its maximum at 72 (the correct number of bits) and that it is almost constant (the values vary between 4 and 0) for a wrong key. We have the same behavior as for the second curve if we try to detect the mark in an unmarked image (Figure 2(Middle) 4 PSNR is the Peak Signal to Noise Ratio [6]. 20 40 60 80 100 120 140 160 200 200 400 600 800 1000 1200 1400 Attempt to detect the message lenght the watermark the number of bits (M) the relative log probability ( P r ) k=180599 k=234567 20 40 60 80 100 120 140 160 4 3.5 3 2.5 2 1.5 1 Attempt to detect the ....

....that the relative log probabilities were below the threshold (generally between 0 and 4) excepting, when we used the correct key and the marked image. 6.3 Stirmark attacks. To test the robustness to different attacks of our algorithm, we applied the program Stirmark 3. 1 5 of Fabien Petitcolas [6] on the marked images. The results are shown in Table 1. Figure 3 shows the behavior of the relative log probability (P r ) in the case of some of the tests. The results show that the algorithm resist generally to the different attacks (excepting the random geometric distortion) Moreover, they ....

M. Kutter and F. A. P. Petitcolas. A fair benchmark for image watermarking systems. In Electronic Imaging '99, Security and Watermarking of Multimedia Contents, volume 3657, pages 219--239, San Jose, CA, USA, January 1999.

....Since an exhaustive search leads to an intractable problem, we demonstrate how a careful pruning of the search space yiels to robust detection of transformations reasonable quickly. The proposed method is evaluated relative to the benchmark series of tests proposed by Kutter and Petitcolas [9] and implemented in the software package Stirmark3 [14] The algorithm performs very well relative to the extensive series of tests implemented in the benchmark. The rest of this paper is structured as follows. In section 2 we describe the embedding approach. Section 3 describes the extraction ....

....power of 2 in order to transform from the spatial domain to DFT and vice versa since we adopt the FFTW package [6] to calculate FFTs of arbitrary size efficiently. 4 Results In this section we evaluate the proposed approach relative to a standard series of tests detailed by Petitcolas and Kutter [15, 9] and then compare the results to the performance of two commercially available algorithms. 9 4.1 Test Results We use the stirmark [14] program to evaluate the algorithm. The tests are divided into the following 8 sections: signal enhancement, compression, scaling, cropping, shearing, rotation, ....

[Article contains additional citation context not shown here]

M. Kutter and F. A. P. Petitcolas. A fair benchmark for image watermarking systems. In Electronic Imaging '99, Security and Watermarking of Multimedia Contents, volume 3657, pages 219--239, San Jose, CA, USA, January 1999.

....1 then compute x 00 i : x i i . X 00 = fx 00 i : 1 i w hg is the marked image. In particular, the embedded marks will survive JPEG compression of X 00 down to quality level q (see Subsection 2. 2 for a more comprehensive robustness assessment) Quality metrics such as the ones in [4, 5] can be used to measure imperceptibility of the mark. If imperceptibility is not satisfactory, then re run the mark embedding algorithm with a higher quality level q. For mark reconstruction, knowledge of the original image and the secret key k is assumed (k is used to regenerate the random ....

M. Kutter and F. A. P. Petitcolas. A fair benchmark for image watermarking systems. In Security and Watermarking of Multimedia Contents, vol. 3657. The Society for Imaging Science and Technology and the International Society for Optical Engineering, San Jose CA, 1999, pp. 226-239.

....Introduction Watermarking methods embed information in a multimedia object in which the modification should be imperceptible. The embedded information can be used as a proof of ownership or as a kind of secret data transmission. Several types of watermarking systems have been proposed since 1990 [1]. Among them, public watermarking is considered to have a broader application value, because it can detect the watermarks without the original object. There are two types of public watermarking systems: one bit watermark and multiple bit watermark [2] The one bit public watermarking system (also ....

M. Kutter and F. A. P. Petitcolas, "A Fair Benchmark for Image Watermarking Systems," SPIE Security and Watermarking of Multimedia Content, San Jose, CA, Jan 1999.

....for this relatively young field include copyright protection and broadcast monitoring. Often, a watermark should be imperceptible: the watermarked and original documents should be perceptually indistinguishable. An attack is any processing of a watermarked document that may impair the watermark [4, 5]. A watermark should also usually be robust, meaning that it cannot be impaired or removed without also making the attacked document useless. Most watermarking methods employ some form of spread spectrum communications [6] whose properties may enable it to meet these requirements. Currently, it ....

M. Kutter, F. A. P. Petitcolas, "A fair benchmark for image watermarking systems," in Proc. SPIE Security and Watermarking of Multimedia Contents, San Jose, CA, USA, Jan. 1999, vol. 3657, pp. 226--239.

No context found.

M. Kutter and F. Petitcolas, "A fair benchmark for image watermarking systems," in Proc. SPIE IS&T/SPIE's 11th Annu. Symp., Electronic Imaging '99: Security and Watermarking of Multimedia Contents, vol. 3657, Jan. 1999.

....schemes. 4.5.5. Security Analysis For watermarking schemes, the security of a scheme is di#cult to prove. Up to now, the best method is to try and attack the watermark with known attacks. In the media domains, the list of attacks are well known and standard benchmark tests like Stirmark [PAK98, KP99] or Checkmark [CHE01] exist. In the code domain, this is not well established. Most of the attacks valid for multimedia are not valid for code. Even if the name is the same, often the meaning is di#erent. For each attack that is relevant for code, we will signal any similar attack in multimedia. ....

M. Kutter and F. Petitcolas. A Fair Benchmark for Image Watermarking Systems. In Security and Watermarking of Multimedia Contents, volume 3657 of Proceedings of the SPIE, pages 226--239, 1999. 95

....as rotation have often been ignored [80] 105] In some cases the watermark is simply said to be robust against common signal processing algorithms and geometric distortions when used on some standard images . This motivated the introduction of a fair benchmark for digital image watermarking in [107]. Similarly, various steganographic systems have shown serious limitations [108] Craver et al. 109] identify at least three kinds of attacks: robustness attacks which aim to diminish or remove the presence of a digital watermark, presentation attacks which modify the content such that the ....

....imperfections typically found in scanners and display devices. Resampling uses the approximating quadratic B spline algorithm [111] An example of these distortions is given in figure 7. StirMark can also perform a default series of tests which serve as a benchmark for image watermarking [107]. Digital watermarking remains a largely untested field and very few authors have published extensive tests on their systems (e.g. 112] A benchmark is needed to highlight promising areas of research by showing which techniques work better than others. One might try to increase the robustness ....

[Article contains additional citation context not shown here]

M. Kutter and F. A. P. Petitcolas, "A fair benchmark for image watermarking systems." In Wong and Delp [106], pp. 226--239, ISBN 0-8194-3128-1.

....as rotation have often been ignored [81] 107] In some cases the watermark is simply said to be robust against common signal processing algorithms and geometric distortions when used on some standard images. This motivated the introduction of a fair benchmark for digital image watermarking in [108]. Similarly, various steganographic systems have shown serious limitations [109] Craver et al. 110] identify at least three kinds of attacks: robustness attacks, which aim to diminish or remove the presence of a digital watermark; presentation attacks, which modify the content such that the ....

....converter imperfections typically found in scanners and display devices. Resampling uses the approximating quadratic B spline algorithm [112] An example of these distortions is given in Fig. 7. StirMark can also perform a default series of tests which serve as a benchmark for image watermarking [108]. Digital watermarking remains a largely untested field and very few authors have published extensive tests on their systems (e.g. 113] A benchmark is needed to highlight promising areas of research by showing which techniques work better than others. One might try to increase the robustness ....

[Article contains additional citation context not shown here]

M. Kutter and F. A. P. Petitcolas, "A fair benchmark for image watermarking systems," in 11th Int. Symp. Electronic Imaging, vol. 3657. San Jose, CA: IS&T and SPIE, Jan. 25--27, 1999.

No context found.

M. Kutter and F. Petitcolas, "A fair benchmark for image watermarking systems," in Security and Watermarking of Multimedia Contents, ser. Proceedings of SPIE, vol. 3657, January 1999, pp. 226--239.

No context found.

M. Kutter and F. A. Petitcolas "A Fair Benchmark for Image Watermarking Systems" Proc. SPIE Security and Watermarking of Multimedia Contents vol. 3657, pp. 226-239, January 1999.

No context found.

M. Kutter and F. A. P. Petitcolas, "A fair benchmark for image watermarking systems," in Proc. SPIE, vol. 3657, San Jose, CA, Jan. 1999, pp. 226--239.

No context found.

Kutter, M. and Petitcolas, F.A.P. (1999): A fair benchmark for image watermarking systems. Proc. of SPIE: Security & Watermarking Multimedia Content, San Jose, CA, USA, 3657, 219-239, Jan. 1999.

No context found.

M. Kutter and F. A. Petitcolas "A Fair Benchmark for Image Watermarking Systems" Proc. SPIE Security and Watermarking of Multimedia Contents vol. 3657, pp. 226-239, January 1999.

No context found.

M. Kutter and F. A. P. Petitcolas. A fair benchmark for image watermarking systems. In Security and Watermarking of Multimedia Contents, vol. 3657. The Society for Imaging Science and Technology and the International Society for Optical Engineering, San JoseCA, 1999, pp. 226-239.

No context found.

M. Kutter and F.A.P. Petitcolas. "A fair benchmark for image watermarking systems", in Proc. of SPIE, Vol. 3657, January 1999.

No context found.

M. Kutter and F. A. P. Petitcolas, "A fair benchmark for image watermarking systems," in Proc. Security and Watermarking of Multimedia Contents, Jan. 1999, pp. 226--239.

No context found.

M. Kutter and F. A. P. Petitcolas, "A fair benchmark for image watermarking systems," in Proc. SPIE Security and Watermarking of Multimedia Contents, vol. 3657, Jan. 1999, pp. 226--239.

No context found.

M. Kutter and F. A. P. Petitcolas, "A fair benchmark for image watermarking systems," in Proc. SPIE Security and Watermarking of Multimedia Contents, Jan. 1999, pp. 226--239.

No context found.

Martin Kutter and Fabien A. P. Petitcolas. A fair benchmark for image watermarking systems. In citeseer.nj.nec.com/article/kutter99fair.html, pages 226--239.

No context found.

M. Kutter and F. A. P. Petitcolas, \A fair benchmark for image watermarking systems,"Proc. of SPIE: Security and Watermarking of Multimedia Contents, Vol. 3657, pp. 226-239, 1999.

No context found.

M. Kutter and F. Petitcolas, \A fair benchmark for image watermarking systems," ########### ## ########## ### #### #### ####, 1999.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC