Developing a robust method of image watermarking that resists rotation, scaling and/or translation (RST) transformations is widely considered to be a more challenging task than developing a method resistant to other attacks. Altering the orientation or size of the image, even only slightly, reduces the receiver¡¦s ability to retrieve the watermark. Protecting against both geometric distortion and signal processing with blind detection is even more problematic. This investigation proposes a novel approach, based on the properties of histograms to measure the numerous global features of all pixels in a cover image and to construct the three-dimensional feature space. The feature space is dynamically partitioned to identify several blocks used to embed the watermark. One feature of the pixels is modified to form a specially distributed histogram for embedding the watermark in a blind digital watermarking method that can be applied to color images. Experimental results demonstrate the robustness of the proposed method against common geometric attacks and signal processing.

This thesis focuses on video watermarking robust against geometric attacks and video compressions. In addition to the requirements for an image watermarking algorithm, a digital video watermarking algorithm has to be robust against advanced video compressions, frame loss, frame swapping, aspect ratio change, frame rate change, intra- and inter-frame filtering, etc. Video compression, especially, the most efficient compression standard, H.264, and geometric attacks, such as rotation and cropping, frame aspect ratio change, and translation, are considered the most challenging attacks for video watermarking algorithms. In this thesis, we first review typical watermarking algorithms robust against geometric attacks and video compressions, and point out their advantages and disadvantages. Then, we propose our robust video watermarking algorithms against Rotation, Scaling and Translation (RST) attacks and MPEG-2 compression based on the logpolar mapping and the phase-only filtering method. Rotation or scaling transformation in the spatial domain results in vertical or horizontal shift in the log-polar mapping (LPM) of the magnitude of the Fourier spectrum of the target frame. Translation has no effect in this domain. This method is very robust to RST attacks and MPEG-2 compression. We also demonstrate that this method can be used as a RST parameters detector to work with other watermarking algorithms to improve their robustness to RST attacks. Furthermore, we propose a new video watermarking algorithm based on the 1D DFT (one-dimensional Discrete Fourier Transform) and 1D projection. This algorithm enhances the robustness to video compression and is able to resist the most advanced video compression, H.264. The 1D DFT for a video sequence along the temporal domain generates an ideal domain, in which the spatial information is still kept and the temporal information is obtained. With detailed analysis and calculation, we choose the frames with highest temporal frequencies to embed the fence-shaped watermark pattern in the Radon transform domain of the selected frames. The performance of the proposed algorithm is evaluated by video compression standards MPEG-2 and H.264; geometric attacks such as rotation, translation, and aspect-ratio changes; and other video processing. The most important advantages of this video watermarking algorithm are its simplicity, practicality and robustness.

This thesis focuses on video watermarking robust against geometric attacks and video compressions. In addition to the requirements for an image watermarking algorithm, a digital video watermarking algorithm has to be robust against advanced video compressions, frame loss, frame swapping, aspect ratio change, frame rate change, intra- and inter-frame filtering, etc. Video compression, especially, the most efficient compression standard, H.264, and geometric attacks, such as rotation and cropping, frame aspect ratio change, and translation, are considered the most challenging attacks for video watermarking algorithms. In this thesis, we first review typical watermarking algorithms robust against geometric attacks and video compressions, and point out their advantages and disadvantages. Then, we propose our robust video watermarking algorithms against Rotation, Scaling and Translation (RST) attacks and MPEG-2 compression based on the logpolar mapping and the phase-only filtering method. Rotation or scaling transformation in the spatial domain results in vertical or horizontal shift in the log-polar mapping (LPM) of the magnitude of the Fourier spectrum of the target frame. Translation has no effect in this domain. This method is very robust to RST attacks and MPEG-2 compression. We also demonstrate that this method can be used as a RST parameters detector to work with other watermarking algorithms to improve their robustness to RST attacks. Furthermore, we propose a new video watermarking algorithm based on the 1D DFT (one-dimensional Discrete Fourier Transform) and 1D projection. This algorithm enhances the robustness to video compression and is able to resist the most advanced video compression, H.264. The 1D DFT for a video sequence along the temporal domain generates an ideal domain, in which the spatial information is still kept and the temporal information is obtained. With detailed analysis and calculation, we choose the frames with highest temporal frequencies to embed the fence-shaped watermark pattern in the Radon transform domain of the selected frames. The performance of the proposed algorithm is evaluated by video compression standards MPEG-2 and H.264; geometric attacks such as rotation, translation, and aspect-ratio changes; and other video processing. The most important advantages of this video watermarking algorithm are its simplicity, practicality and robustness.

Σκοπός της συγκεκριμένης εργασίας είναι η παρουσίαση των σημαντικότερων τεχνικών υδατογράφησης (watermarking techniques) ψηφιακών εικόνων με ανθεκτικότητα σε γεωμετρικούς μετασχηματισμούς και η περιγραφή και υλοποίηση ενός τέτοιου συστήματος. Οι γεωμετρικοί μετασχηματισμοί, στους οποίους δόθηκε περισσότερη έμφαση κατά την υλοποίηση του συστήματος υδατογράφησης ψηφιακών εικόνων που θα παρουσιαστεί παρακάτω, έχουν να κάνουν με την κλιμάκωση (scaling), την περιστροφή (rotation) και τον μετασχηματισμό συντεταγμένων (translation) μιας εικόνας καθώς επίσης και με συνδυασμούς αυτών. Παρόλα αυτά όμως το σύστημα που περιγράφεται αντιμετωπίζει σε ικανοποιητικό βαθμό και άλλες γεωμετρικές επιθέσεις (όπως shearing,aspect ratio change, projective, template remove κ.α.).Η ανθεκτικότητα ενός συστήματος υδατογράφησης σε αυτούς τους μετασχηματισμούς έχει να κάνει με την διατήρηση του υδατογραφήματος και την ικανοποιητική ανίχνευση του σε εικόνες που έχουν υποστεί τέτοιου είδους μετασχηματισμούς. Στην εργασία, αρχικά γίνεται μια εκτενής αναφορά στην τεχνική της υδατογράφησης και τις εφαρμογές της και έπειτα τονίζονται τα σημαντικότερα χαρακτηριστικά των συστημάτων υδατογράφησης ψηφιακού υλικού. Στη συνέχεια, ορίζονται οι γεωμετρικές επιθέσεις σε εικόνες και παρουσιάζονται οι σημαντικότερες μέθοδοι υδατογράφησης ψηφιακών εικόνων, που υπάρχουν στην βιβλιογραφία, με ανθεκτικότητα στις επιθέσεις αυτές. Στην εργασία αυτή υλοποιήθηκε μία από αυτές τις μεθόδους και έγινε προσπάθεια για επέκταση και βελτίωση της σε ορισμένα σημεία που υστερούσε. Η αρχική και βελτιωμένη μέθοδος περιγράφονται αναλυτικά. Στο τέλος ένας αριθμός πειραματικών αποτελεσμάτων αποδεικνύουν την ανθεκτικότητα του νέου βελτιωμένου συστήματος που υλοποιήθηκε σε ένα μεγάλο πλήθος γεωμετρικών επιθέσεων. / The aim of this work is the presentation of the most important watermarking techniques for digital images with robustness to geometric transformations and the description and implementation of such a system. The geometric transformations, to which was given more emphasis at the implementation of the digital images watermarking system, that will be presented below, are the scaling, the rotation and the transformation of coordinates (translation) of an image as well as the combinations of these. Nevertheless, the system that is described faces in satisfactory degree and the other geometric attacks (as the shearing, the aspect ratio change, the projective, the template remove etc.). The robustness of a watermarking system to these transformations it is simply translated as the maintenance of the watermark and its satisfactory detection in images that have existed such type of transformations. In this work, it initially becomes a extensive report to the watermark technique and to its applications and then the more important characteristics of digital material watermarking systems are highlighted. Afterwards, the geometric attacks of digit images are determined and the most important watermarking methods of digital images with robustness to these attacks, that exist in the bibliography, are presented. In this work, one of these methods was implemented and it became some effort for its extension and its improvement in certain points that disadvantages. The initial and improved methods are analytically described. At the end, a number of experimental results are presented, that they prove the robustness of the new improved system to a large number of geometric attacks.

This thesis focuses on video watermarking robust against geometric attacks and video compressions. In addition to the requirements for an image watermarking algorithm, a digital video watermarking algorithm has to be robust against advanced video compressions, frame loss, frame swapping, aspect ratio change, frame rate change, intra- and inter-frame filtering, etc. Video compression, especially, the most efficient compression standard, H.264, and geometric attacks, such as rotation and cropping, frame aspect ratio change, and translation, are considered the most challenging attacks for video watermarking algorithms. In this thesis, we first review typical watermarking algorithms robust against geometric attacks and video compressions, and point out their advantages and disadvantages. Then, we propose our robust video watermarking algorithms against Rotation, Scaling and Translation (RST) attacks and MPEG-2 compression based on the logpolar mapping and the phase-only filtering method. Rotation or scaling transformation in the spatial domain results in vertical or horizontal shift in the log-polar mapping (LPM) of the magnitude of the Fourier spectrum of the target frame. Translation has no effect in this domain. This method is very robust to RST attacks and MPEG-2 compression. We also demonstrate that this method can be used as a RST parameters detector to work with other watermarking algorithms to improve their robustness to RST attacks. Furthermore, we propose a new video watermarking algorithm based on the 1D DFT (one-dimensional Discrete Fourier Transform) and 1D projection. This algorithm enhances the robustness to video compression and is able to resist the most advanced video compression, H.264. The 1D DFT for a video sequence along the temporal domain generates an ideal domain, in which the spatial information is still kept and the temporal information is obtained. With detailed analysis and calculation, we choose the frames with highest temporal frequencies to embed the fence-shaped watermark pattern in the Radon transform domain of the selected frames. The performance of the proposed algorithm is evaluated by video compression standards MPEG-2 and H.264; geometric attacks such as rotation, translation, and aspect-ratio changes; and other video processing. The most important advantages of this video watermarking algorithm are its simplicity, practicality and robustness.

This thesis focuses on video watermarking robust against geometric attacks and video compressions. In addition to the requirements for an image watermarking algorithm, a digital video watermarking algorithm has to be robust against advanced video compressions, frame loss, frame swapping, aspect ratio change, frame rate change, intra- and inter-frame filtering, etc. Video compression, especially, the most efficient compression standard, H.264, and geometric attacks, such as rotation and cropping, frame aspect ratio change, and translation, are considered the most challenging attacks for video watermarking algorithms. In this thesis, we first review typical watermarking algorithms robust against geometric attacks and video compressions, and point out their advantages and disadvantages. Then, we propose our robust video watermarking algorithms against Rotation, Scaling and Translation (RST) attacks and MPEG-2 compression based on the logpolar mapping and the phase-only filtering method. Rotation or scaling transformation in the spatial domain results in vertical or horizontal shift in the log-polar mapping (LPM) of the magnitude of the Fourier spectrum of the target frame. Translation has no effect in this domain. This method is very robust to RST attacks and MPEG-2 compression. We also demonstrate that this method can be used as a RST parameters detector to work with other watermarking algorithms to improve their robustness to RST attacks. Furthermore, we propose a new video watermarking algorithm based on the 1D DFT (one-dimensional Discrete Fourier Transform) and 1D projection. This algorithm enhances the robustness to video compression and is able to resist the most advanced video compression, H.264. The 1D DFT for a video sequence along the temporal domain generates an ideal domain, in which the spatial information is still kept and the temporal information is obtained. With detailed analysis and calculation, we choose the frames with highest temporal frequencies to embed the fence-shaped watermark pattern in the Radon transform domain of the selected frames. The performance of the proposed algorithm is evaluated by video compression standards MPEG-2 and H.264; geometric attacks such as rotation, translation, and aspect-ratio changes; and other video processing. The most important advantages of this video watermarking algorithm are its simplicity, practicality and robustness.

Abstract During the last few decades, digital watermarking techniques have gained a lot of interest. Such techniques enable hiding imperceptible information to images; information which can be extracted later from those images. As a result, digital watermarking techniques have many interesting applications for example in Internet distribution. Contents such as images are today manipulated mainly in digital form; thus, traditionally, the focus of watermarking research has been the digital domain. However, a vast amount of images will still appear in some physical format such as in books, posters or labels, and there are a number of possible applications of hidden information also in image printouts. In this case, an additional level of challenge is introduced, as the watermarking technique should be robust to extraction from printed output. In this thesis, methods are developed, where a watermarked image appears in a printout and the invisible information can be later extracted using a scanner or mobile phone camera and watermark extraction software. In these cases, the watermarking method has to be carefully designed because both the printing and capturing process cause distortions that make watermark extraction challenging. The focus of the study is on developing blind, multibit watermarking techniques, where the robustness of the algorithms is tested in an office environment, using standard office equipment. The possible effect of the background of the printed images, as well as compound attacks, are both paid particular attention to, since these are considered important in practical applications. The main objective is thus to provide technical means to achieve high robustness and to develop watermarking methods robust to printing and scanning process. A secondary objective is to develop methods where the extraction is possible with the aid of a mobile phone camera. The main contributions of the thesis are: (1) Methods to increase watermark extraction robustness with perceptual weighting; (2) Methods to robustly synchronize the extraction of a multibit message from a printout; (3) A method to encode a multibit message, utilizing directed periodic patterns and a method to decode the message after attacks; (4) A demonstrator of an interactive poster application and a key based robust and secure identification method from a printout. / Tiivistelmä Digitaalinen vesileimaus on parin viime vuosikymmenen aikana runsaasti huomiota saanut tekniikka, jonka avulla kuviin voidaan piilottaa aistein havaitsematonta tietoa. Tämä tieto voidaan myöhemmin poimia esiin, minkä vuoksi sovelluskohteita esimerkiksi Internetin kautta tapahtuvassa jakelussa on useita. Perinteisesti vesileimaustekniikat keskittyvät pelkästään digitaalisessa muodossa pysyvään tietoon. Kuitenkin iso osa kuvainformaatiosta saa yhä vielä myös fyysisen muodon esimerkiksi kirjoissa, julisteissa ja etiketeissä. Myös vesileimauksella on useita sovelluskohteita painettujen kuvienkin osalta. Vesileimausta ajatellen painatus tuo kumminkin omat erityishaasteensa vesileimaustekniikoille. Tässä väitöskirjassa kehitetään menetelmiä, jotka mahdollistavat piilotetun tiedon säilymisen painetussa kuvassa. Piilotettu tieto voidaan lukea käyttämällä skanneria tai matkapuhelimen kameraa tiedon digitalisointiin. Digitalisoinnin jälkeen vesileimausohjelma osaa lukea piilotetun tiedon. Vesileimauksen osalta haasteellisuus tulee vääristymistä, joita sekä kuvien tulostus sekä digitalisointi aiheuttavat. Väitöstyössä keskitytään monibittisiin vesileimaustekniikoihin, joissa alkuperäistä kuvaa ei tarvita vesileimaa poimittaessa. Väitöstyössä kehitetyt menetelmät on testattu toimistoympäristössä standardi toimistolaitteita käyttäen. Käytännön sovelluksia ajatellen, testeissä on kiinnitetty huomiota myös yhdistelmähyökkäysten sekä painetun kuvan taustan vaikutukseen algoritmin robustisuudelle. Ensisijainen tavoite on kehittää menetelmiä, jotka kestävät printtaus ja skannaus operaation. Toinen tavoite on tiedon kestävyys luettaessa tietoa matkapuhelimen kameran avulla. Väitöskirjassa tarkastellaan ja kehitellään ratkaisuja neljälle eri osa-alueelle: (1) Ihmisaisteja mallintavien menetelmien käyttö vesileimauksen kestävyyden lisäämiseksi; (2) Robusti synkronointi luettaessa monibittistä tietoa painotuotteesta; (3) Suunnattuja jaksollisia kuvioita käyttävä menetelmä, joka mahdollistaa monibittisen tiedon koodaamisen ja dekoodaamisen hyökkäysten jälkeen; (4) Sovellustasolla tarkastellaan kahta pääsovellusta: interaktiivinen juliste sekä kestävä ja turvattu avaimen avulla tapahtuva painotuotteen identifiointi.