Current content protection systems rely primarily on applied cryptographic techniques but there is an increased use of
forensic solutions in images, music and video distribution alike. The two approaches differ significantly, both in terms
of technology and in terms of strategy, and thus it begs the question: will one approach take over in the long run, and if
so which one? Discussing the evolution of both cryptographic and forensic solutions, we conclude that neither approach
is ideal for all constituents, and that in the video space at least they will continue to co-exist for the foreseeable future -
even if this may not be the case for other media types. We also analyze shortcomings of these approaches, and suggest
that new solutions are necessary in this still emerging marketplace.
Proc. SPIE. 5020, Security and Watermarking of Multimedia Contents V
KEYWORDS: Signal to noise ratio, Sensors, Interference (communication), Distortion, Linear filtering, Digital watermarking, Signal processing, Distance measurement, Electronic filtering, Signal detection
Most watermarking applications require that the embedded watermark be
imperceptible. Accordingly, perceptual masking models that identify
unperceived regions of the signal were adapted in a straightforward
manner to watermarking. The derived mask -- or slack -- is often
interpreted as the maximal allowed distortion within a given region
of the signal; it is used in many watermarking embedding methods to
shape a white spectrum message, in the relevant transform domain
(space, frequency). Such a usage of the mask is intuitively
satisfying since imperceptibility is indeed guaranteed; yet, it
discards any guarantee of robustness to attacks -- another fundamental, necessary property of watermarks. The trade-off between fidelity and robustness has been little addressed so far due in great part to the absence of an accurate measure of perceptual distortion. In this paper we study this trade-off using Watson's measure of perceptual distance between two images as the measure of fidelity. Based on a constrained perceptual distance, the embedder must maximize the watermark's robustness while assuming a knowledgeable attacker will attempt to remove the watermark. Solving this problem leads to an optimized watermark strength for each location of the content.
Masking models are mostly used in data compression algorithms and serve to shape the quantization noise. They were introduced in watermarking as to indicate the regions where the watermark could be introduced without perceptible artifacts. This allowed to embed more watermark energy, for a given absolute distortion constraint, than if no mask is used. Yet, little attention has been paid to the consequences of using these masks with respect to detection performance. In this work, it is shown that blind use of masking models facilitates the attacker's role, and eventually results in severe decreases of detection statistic at the detector, even for reasonable attack distortions.
The two main objectives of this paper are: (1) to better define the public-key (PK) watermarking problem -- in terms of properties, design requirements and usage, and (2) to propose one solution to the problem by using neural networks functions. Our survey of public key watermarking begins with the review of the state of the art. Different aspects of PK watermarking are then discussed, among which: basic robustness properties, usage of PK systems, attacks on the public and secret detectors, types of PK strategies, and strong vs weak PK watermarking systems. Accordingly, a PK system using multi-layers neural networks (NN) functions is proposed to match many PK system requirements. The approach is shortly presented for the linear case. Theoretical results are given, showing that it is possible to design PK systems approaching the detection performances of secret key watermarking-- a very unusual feature of PK systems. Experimental results are given on both simulated signals and image, confirming the predicted results and showing great resistance to JPEG compression. The paper ends with openings for new research directions.
This paper presents a transform domain allocation method for watermarking of digital multimedia content; specific analytical results are obtained for audio watermarking. To solve the problem, the watermarking transmitter, channel and receiver are formally defined and watermarking is described in a detection theory framework. Describing the channel by a generic attack model and using detection theory, the challenge is to minimize the probability of detection error at the receiver. Using strong assumptions on the original and watermark signals to derive analytical results, an optimal allocation method is deduced. The allocation strategy is compared to strategies in most present systems. Moreover, the assumptions on the signals reveal to be realistic in a given implementation of the watermarking system. The allocation method is being patented for digital multimedia content.