In the immediate future, multimedia product distribution through the Internet will become main stream. However, it can also have the side effect of unauthorized duplication and distribution of multimedia products. That effect could be a critical challenge to the legal ownership of copyright and intellectual property. Many schemes have been proposed to address these issues; one is digital watermarking which is appropriate for image and video copyright protection. Videos distributed via the Internet must be processed by compression for low bit rate, due to bandwidth limitations. The most widely adapted video compression standard is MPEG-4. Discrete cosine transform (DCT) domain watermarking is a secure algorithm which could survive video compression procedures and, most importantly, attacks attempting to remove the watermark, with a visibly degraded video quality result after the watermark attacks. For a commercial broadcasting video system, real-time response is always required. For this reason, an FPGA hardware implementation is studied in this work. This thesis deals with video compression, watermarking algorithms and their hardware implementation with FPGAs. A prototyping VLSI architecture will implement video compression and watermarking algorithms with the FPGA. The prototype is evaluated with video and watermarking quality metrics. Finally, it is seen that the video qualities of the watermarking at the uncompressed vs. the compressed domain are only 1dB of PSNR lower. However, the cost of compressed domain watermarking is the complexity of drift compensation for canceling the drifting effect.

En traçage de traîtres, plusieurs copies d'un même contenu sont personnalisées avant distribution pour identifier d'éventuelles fuites d'informations. Dans le contexte de la distribution de contenus multimédia, cette personnalisation repose sur une construction jointe d'un code composé d'identifiants uniques et d'un algorithme de dissimulation d'informations. Une des problématiques majeure de cette construction est de conserver des propriétés d'identification même en cas d'attaques par collusion d'une coalition de traîtres, c'est-à-dire lorsque plusieurs copies d'utilisateurs (les traîtres) sont utilisées pour créer une copie pirate inconnue du système. Ces attaques doivent être contrées d'une part, au niveau du code traçant et d'autre part au niveau de l'algorithme d'insertion. Pour une distribution de contenus à grande échelle, il est nécessaire d'aborder le problème de la génération des identifiants de manière probabiliste. Le code de Gabor Tardos est alors l'approche la plus performante pour un alphabet binaire. Cependant, pour des identifiants à valeurs réelles respectant une contrainte de distorsion à l'insertion, ces identifiants binaires ne sont plus optimaux en terme d'erreur de décodage. Dans cette thèse, nous avons proposé un décodeur MAP non informé pour améliorer le décodage du code de Tardos. Nous nous sommes aussi intéressés à combiner les approches de codes à valeurs binaires et réelles afin d'améliorer les performances de traçage. Enfin, nous avons abordé le problème de la complexité de décodage avec une approche hiérarchique de codes à deux niveaux. / In Traitor Tracing, preventing leaks of a digital content comes with the personalization of each delivered copy. For multimedia content distribution, such as video on demand applications, the personalization depends on a joint construction between a code of unique sequences and a data hiding tool. The major issue is to keep tracing properties even in case of collusion attacks proceed by a coalition of traitors. Collusion attacks have to be counteract both on the code layer with anti-collusion construction and on the data hiding layer with robust watermarking. For large scale multimedia content distribution, it is necessary to tackle the problem of the code generation in a probabilistic manner. The binary code of Gabor Tardos is actually the best approach of probabilistic traitor tracing codes. However, for real valued signals prone to distortion constraint, which is the context in multimedia digital watermarking, Tardos binary fingerprints are no longer optimal in term of decoding error. Our work concerns several aspects of probabilistic codes construction. Firstly, we gives new accusation functions in the Tardos tracing framework with a non informed MAP (Maximum A Posteriori) decoder leading to decrease errors probabilities. Secondly, we proposed a joint construction of binary and real values codes so as to improve collusion robustness. Finally we focus on a hierarchical two layers codes construction which reduces the Tardos decoding complexity.

In this thesis, we present a novel, elegant and simple method for secure transaction authentication and non-repudiation for trading multimedia content. Multimedia content can be video, images, text documents, music, or any form of digital signal, however here we will focus particular on still images with application to video. We will provide proof that not only can receiving parties within a transaction be untrustworthy, but the owner, or members within an owning party, also cannot be trusted. Known as the insider attack, this attack is particularly prevalent in multimedia transactions. Thus the focus of the thesis is on the prevention of piracy, with particular emphasis on the case where the owner of a document is assumed to be capable of deceit, placing the system under the assumption of mutual distrust. We will introduce a concept called staining, which will be used to achieve authentication and non-repudiation. Staining is composed of two key components: (1) public-key cryptography; and (2) steganographic watermarking. The idea is to watermark a multimedia document after encryption, thereby introducing a stain on the watermark. This stain is due to the non-commutative nature of the scheme, so that decryption will be imperfect, leaving a residue of the cryptographic process upon the watermark. Essentially, secrets from the owner (the watermark) and the receiver (the cryptographic key) are entangled rather than shared, as in most schemes. We then demonstrate our method using image content and will test several different common cryptographic systems with a spread-spectrum type watermark. Watermarking and cryptography are not usually combined in such a manner, due to several issues such as the rigid nature of cryptography. Contrary to the expectation that there will be severe distortions caused to the original document, we show that such an entanglement is possible without destroying the document under protection. We will then attack the most promising combination of systems by introducing geometric distortions such as rotation and cropping, as well as compressing the marked document, to demonstrate that such a method is robust to typical attacks. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1297339 / Thesis (Ph.D.) - University of Adelaide, School of Electrical and Electronic Engineering, 2007

Thesis (PhD)--Macquarie University, Division of Information and Communication Sciences, Department of Computing, 2008. / Bibliography: p. 144-153. / Introduction -- Background -- Overview of watermarking -- Natural language watermarking -- Software watermarking -- Semi-blind and reversible database watermarking -- Blind and reversible database watermarking -- Conclusion and future research -- Bibliography. / Digital watermarking has generated significant research and commercial interest in the past decade. The primary factors contributing to this surge are widespread use of the Internet with improved bandwidth and speed, regional copyright loopholes in terms of legislation; and seamless distribution of multimedia content due to peer-to-peer file-sharing applications. -- Digital watermarking addresses the issue of establishing ownership over mul-timedia content through embedding a watermark inside the object. Ideally, this watermark should be detectable and/or extractable, survive attacks such as digital reproduction and content-specific manipulations such as re-sizing in the case of images, and be invisible to the end-user so that the quality of the content is not degraded significantly. During detection or extraction, the only requirements should be the secret key and the watermarked multimedia object, and not the original un-marked object or the watermark inserted. Watermarking scheme that facilitate this requirement are categorized as blind. In recent times, reversibility of watermark has also become an important criterion. This is due to the fact that reversible watermarking schemes can provided security against secondary watermarking attacks by using backtracking algorithms to identify the rightful owner. A watermarking scheme is said to be reversible if the original unmarked object can be regenerated from the watermarked copy and the secret key. / This research covers three multimedia content types: natural language documents, software, and databases; and discusses the current watermarking scenario, challenges, and our contribution to the field. We have designed and implemented a natural language watermarking scheme that uses the redundancies in natural languages. As a result, it is robust against general attacks against text watermarks. It offers additional strength to the scheme by localizing the attack to the modified section and using error correction codes to detect the watermark. Our first contribution in software watermarking is identification and exploitation of weaknesses in branch-based software watermarking scheme proposed in [71] and the software watermarking algorithm we present is an improvised version of the existing watermarking schemes from [71]. Our scheme survives automated debugging attacks against which the current schemes are vulnerable, and is also secure against other software-specific attacks. We have proposed two database watermarking schemes that are both reversible and therefore resilient against secondary watermarking attacks. The first of these database watermarking schemes is semi-blind and requires the bits modified during the insertion algorithm to detect the watermark. The second scheme is an upgraded version that is blind and therefore does not require anything except a secret key and the watermarked relation. The watermark has a 89% probability of survival even when almost half of the data is manipulated. The watermarked data in this case is extremely useful from the users' perspective, since query results are preserved (i.e., the watermarked data gives the same results for a query as the nmarked data). -- The watermarking models we have proposed provide greater security against sophisticated attacks in different domains while providing sufficient watermark-carrying capacity at the same time. The false-positives are extremely low in all the models, thereby making accidental detection of watermark in a random object almost negligible. Reversibility has been facilitated in the later watermarking algorithms and is a solution to the secondary watermarking attacks. We shall address reversibility as a key issue in our future research, along with robustness, low false-positives and high capacity. / Mode of access: World Wide Web. / xxiv, 156 p. ill. (some col.)

Thesis (M.Eng. (Electrical, Electromic and Computer Engineering)) -- University of Pretoria, 2007. / Includes bibliographical references (leaves 82-87)

Thesis (MEng)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: We design and implement a prototype digital watermarking scheme that uses spread-transform dither modulation to embed hidden binary message signals into uncompressed digital video signals. The hidden messages are intended to identify the original recipient of a particular copy of a digital video, so that in the event that a pirate copy of that video is found, the recipient can be incriminated as a pirate. We test the ability of the watermark signal to survive various modifications to the watermarked video, including lossy compression, frame deletion, and amplitude scaling. We find that the scheme performs well when watermarked videos remain uncompressed, but the scheme is not sufficiently robust against lossy compression. We also implement a dynamic traitor tracing scheme based on the scheme of Fiat and Tassa, which is designed to identify a group of pirates who combine their watermarked videos in an attempt to defeat the watermarking scheme. We test the traitor tracing scheme in a simulated digitally watermarked video broadcast system where a pirate rebroadcast is monitored. We find that when the pirate rebroadcast is delayed, our traitor tracing scheme identifies and disconnects the pirates more quickly than the scheme of Fiat and Tassa. / AFRIKAANSE OPSOMMING: Ons ontwerp en implementeer ’n digitale watermerkskema-prototipe wat strektransform-bibbermodulasie gebruik om binêre boodskapseine in digitale videoseine te verskuil. Die doel van die versteekte boodskappe is om die oorspronklike ontvanger van ’n bepaalde digitale video-afskrif te identifiseer. Indien ’n onwettige afskrif van die video ontdek word, kan die oorspronklike ontvanger daarvoor verantwoordelik gehou word. Ons toets die vermoë van die watermerksein om verskeie veranderinge aan die gewatermerkte video te oorleef, byvoorbeeld kompressie, die verwydering van videorame, en amplitudeskalering. Die skema vaar goed in die afwesigheid van videokompressie, maar is nie robuus wanneer die video beduidend saamgepers word nie. Ons implementeer ook ’n dinamiese skema vir verraaier-opsporing, gebaseer op ’n tegniek deur Fiat en Tassa. Die doel van hierdie skema is om groepe gebruikers te identifiseer wat saamwerk om ’n watermerkskema te kul deur gewatermerkte videos te kombineer. Ons toets die verraaier-opsporingskema in ’n simulasie van ’n gewatermerkte video-uitsendingstelsel wat ’n roofuitsending monitor. Wanneer die roofuitsending vertraag is, vind ons dat die voorgestelde verraaier-opsporingskema die skuldige gebruikers vinniger uitwys as die skema van Fiat en Tassa.

Orientador: Alexandre César Rodrigues da Silva / Resumo: Com a crescente evolução dos equipamentos eletrônicos, muitos dados digitais têm sido produzidos, copiados e distribuídos com facilidade, gerando uma grande preocupação com a sua segurança. Dentre as várias técnicas usadas para proteger os dados digitais, tem-se as técnicas de inserção e extração de marca d'água em imagens digitais. Uma marca d'água pode ser qualquer informação como, por exemplo, um código, um logotipo, ou uma sequência aleatória de letras e números, que vise garantir a autenticidade e a proteção dos direitos autoriais dos dados. Neste trabalho, estudou-se sobre as técnicas existentes de inserção e extração de marca d'água digital, abordando desde seu conceito até o desenvolvimento de algoritmos de inserção e extração de marca d'água em imagens digitais. Desenvolveu-se um método para aumentar a robustez de técnicas de marca d' água digital pela decomposição da imagem em duas partes: estrutural (áreas homogêneas) e de detalhes (áreas com ruídos, texturas e bordas). Contudo, a marca d'água é inserida na parte de detalhes por se tratar de áreas menos afetadas por operações de processamento digital de imagens. Os resultados mostraram que o método proposto aumentou a robustez das técnicas da marca d'água testadas. Baseado nos resultados obtidos, desenvolveu-se uma nova técnica de marca d'água digital, utilizando a transformada discreta de wavelets, a decomposição de imagens e a transformada discreta do cosseno. / Abstract: With the increasing evolution of technological equipment, many digital data have been easily produced, copied and distributed by generating a great concern for their security. Among the various techniques used to protect the digital data, there are techniques for inserting and extracting a watermark into digital images. A watermark can be any information, such as a code, a logo, or a random sequence of letters and numbers, aiming to ensure the authenticity and copyright protection. In this work are studied, existing insertion and extraction techniques in digital watermarking, by covering from its concept to the development of watermark insertion and extraction algorithms, in digital images. A method was developed to increase the robustness of digital watermarking techniques by decomposing the image into two parts: structural (homogeneous areas) and details (areas with noises, textures and edges). However, the watermark is inserted in the detail area due to be less affected areas by digital image processing. The results showed that the proposed method increased the robustness of the tested watermarking techniques. Based on the results obtained, we developed a new digital watermark technique using discrete wavelet transform, image decomposition and discrete cosine transform. / Doutor

Submitted by Ivan Oliveira Lopes (io.lopes@ifsp.edu.br) on 2018-09-22T02:34:15Z No. of bitstreams: 1 MARDI_Marca dagua Robusta via Decomposicao de Imagens.pdf: 69551246 bytes, checksum: 895274cb196d7f5075a7bf92bba60e9f (MD5) / Approved for entry into archive by Cristina Alexandra de Godoy null (cristina@adm.feis.unesp.br) on 2018-09-25T19:59:12Z (GMT) No. of bitstreams: 1 lopes_io_dr_ilha.pdf: 69617454 bytes, checksum: 7a93538756770bc0935aa4b85b95f4b7 (MD5) / Made available in DSpace on 2018-09-25T19:59:12Z (GMT). No. of bitstreams: 1 lopes_io_dr_ilha.pdf: 69617454 bytes, checksum: 7a93538756770bc0935aa4b85b95f4b7 (MD5) Previous issue date: 2018-07-27 / Com a crescente evolução dos equipamentos eletrônicos, muitos dados digitais têm sido produzidos, copiados e distribuídos com facilidade, gerando uma grande preocupação com a sua segurança. Dentre as várias técnicas usadas para proteger os dados digitais, tem-se as técnicas de inserção e extração de marca d'água em imagens digitais. Uma marca d'água pode ser qualquer informação como, por exemplo, um código, um logotipo, ou uma sequência aleatória de letras e números, que vise garantir a autenticidade e a proteção dos direitos autoriais dos dados. Neste trabalho, estudou-se sobre as técnicas existentes de inserção e extração de marca d'água digital, abordando desde seu conceito até o desenvolvimento de algoritmos de inserção e extração de marca d'água em imagens digitais. Desenvolveu-se um método para aumentar a robustez de técnicas de marca d' água digital pela decomposição da imagem em duas partes: estrutural (áreas homogêneas) e de detalhes (áreas com ruídos, texturas e bordas). Contudo, a marca d'água é inserida na parte de detalhes por se tratar de áreas menos afetadas por operações de processamento digital de imagens. Os resultados mostraram que o método proposto aumentou a robustez das técnicas da marca d'água testadas. Baseado nos resultados obtidos, desenvolveu-se uma nova técnica de marca d'água digital, utilizando a transformada discreta de wavelets, a decomposição de imagens e a transformada discreta do cosseno. / With the increasing evolution of technological equipment, many digital data have been easily produced, copied and distributed by generating a great concern for their security. Among the various techniques used to protect the digital data, there are techniques for inserting and extracting a watermark into digital images. A watermark can be any information, such as a code, a logo, or a random sequence of letters and numbers, aiming to ensure the authenticity and copyright protection. In this work are studied, existing insertion and extraction techniques in digital watermarking, by covering from its concept to the development of watermark insertion and extraction algorithms, in digital images. A method was developed to increase the robustness of digital watermarking techniques by decomposing the image into two parts: structural (homogeneous areas) and details (areas with noises, textures and edges). However, the watermark is inserted in the detail area due to be less affected areas by digital image processing. The results showed that the proposed method increased the robustness of the tested watermarking techniques. Based on the results obtained, we developed a new digital watermark technique using discrete wavelet transform, image decomposition and discrete cosine transform.

The aim of this dissertation is to develop a new algorithm to embed a watermark in JPEG compressed images, using encoding methods. This encompasses the embedding of proprietary information, such as identity and authentication bitstrings, into the compressed material. This watermark encoding scheme involves combining entropy coding with homophonic coding, in order to embed a watermark in a JPEG image. Arithmetic coding was used as the entropy encoder for this scheme. It is often desired to obtain a robust digital watermarking method that does not distort the digital image, even if this implies that the image is slightly expanded in size before final compression. In this dissertation an algorithm that combines homophonic and arithmetic coding for JPEG images was developed and implemented in software. A detailed analysis of this algorithm is given and the compression (in number of bits) obtained when using the newly developed algorithm (homophonic and arithmetic coding). This research shows that homophonic coding can be used to embed a watermark in a JPEG image by using the watermark information for the selection of the homophones. The proposed algorithm can thus be viewed as a ‘key-less’ encryption technique, where an external bitstring is used as a ‘key’ and is embedded intrinsically into the message stream. The algorithm has achieved to create JPEG images with minimal distortion, with Peak Signal to Noise Ratios (PSNR) of above 35dB. The resulting increase in the entropy of the file is within the expected 2 bits per symbol. This research endeavor consequently provides a unique watermarking technique for images compressed using the JPEG standard. / Dissertation (MEng)--University of Pretoria, 2008. / Electrical, Electronic and Computer Engineering / unrestricted

Le tatouage d’images (ou watermarking) consiste à insérer de l’information de copyright, ou d’autres messages de vérification, invisible à l’oeil humain. Parmi toutes les approches possibles, le tatouage psychovisuel est le plus performant. Il consiste à utiliser les connaissances du Système Visuel Humain (SVH) pour en déduire la force du tatouage de manière optimale. Dans cette étude, nous avons proposé une nouvelle méthode de tatouage d'image psychovisuel qui combine les avantages des ondelettes à celle du domaine spatial par l’utilisation d'une représentation en ondelettes à échelle mixée (JNDEM). En considérant la densité des coefficients d'ondelettes dominants, notre système est capable de différencier les zones uniformes, celles incluant des contours, et enfin celles texturées. Nous avons sélectionné les méthodes efficaces de tatouage psychovisuel adaptée à chaque zone, à savoir celle de Chou pour les zones uniformes, de nouveau celle de Chou pour les contours, et finalement celle de Qi pour les textures. Nous avons aussi pris en compte la sensibilité du SVH qui varie avec l’orientation locale de l’image. Nous avons comparé notre JNDEM à quatre autres méthodes de tatouage de la littérature en termes de qualité visuelle et de robustesse. La méthode JNDEM conduit à une grande qualité visuelle mesurée par des tests objectifs et subjectifs. De plus elle est robuste aux attaques. Ce dernier point est d'une grande importance pour les applications réelles. En conclusion, la méthode proposée est le meilleur compromis entre la qualité visuelle et la résistance aux attaques comparée aux quatre autres méthodes testées. / Watermarking consists in inserting copyright information, or other verification messages, invisible to the human eye. Among all the possible approaches, perceptual watermarking is the most efficient. Perceptual image watermarking consists in using knowledge of the Human Visual System (HVS) to choose the strength of the watermark. In this study, we propose a new perceptual image watermarking method (JNDEM ) that combines the advantages of both the wavelet domain and the spatial domain since a mixed-scale wavelet representation is applied. By considering the density of the dominant wavelet coefficients, our scheme is able to differentiate uniform, edge and texture areas. We selected effective Just Noticeable Difference (JND) models from the literature, i.e. luminance and edge masking developed by Chou, and texture masking developed by Qi. We also took into account the HVS sensitivity which varies with the orientation of the image activity. We have compared our JNDEM to four other JND models from the literature in terms of visual quality and of robustness to attacks. Results show that the proposed method yields a high visual quality and is very robust against attacks. This last point is of great importance for real applications. In conclusion, the proposed method is the best trade-off between visual quality and resistance to attacks among the four other tested methods.