DoctorateOpen Access

Discrete wavelet transform discrete cossine transform based digital watermarking scheme for still images and video signals

2006
0 views
0 downloads
Advisor: Prof. Dr. Metin Yücel

Abstract (EN)

There has been an explosion in the use and distribution of multimedia content because of therapid development of information technologies for multimedia services. The success of theInternet, cost-effective and popular digital recording and storage devices, and the promise ofhigher bandwidth and quality of services for both wired and wireless networks have made itpossible to easily create, replicate, transmit, and distribute digital content. Consequently, thishas led to a strong demand for reliable and secure copyright protection techniques formultimedia data. In digital watermarking methods, invisible or inaudible data is embedded tomultimedia content. The hidden data for such purposes is called a watermark and it containscopyright information. Digital watermarking methods can be used for copyright protection,fingerprint trace, copy protection, broadcast monitoring, data authentication and data hidingpurposes. Although watermarking algorithms have different characteristic feature,imperceptibility, robustness, security, and payload are most important features for effectivewatermarking.In this work, we propose a new digital watermark method using the DWT prior to the DCT toprovide better imperceptibility in harmony with the human visual system, and higherrobustness against signal processing attacks. The robustness is improved by the achievedability to increase the energy of the watermark. The energy is increased locally by exploitingthe properties of the HVS such that the human eye will not notice it. DWT is used as acomputationally efficient version of the frequency models for HVS. In this way,imperceptibility is also improved due to the DWT. DCT is also improved robustness againstto signal processing attacks.We start the watermarking process by applying the DWT to the original image, andsubsequently applying the DCT to the DWT sub-bands. Firstly, we apply the method to stillimages that are most used in the literature, then generalize for MPEG-2 and MPEG-4 videostreams. But the method is most suitable for DWT based system such as JPEG2000 stillimage compression standart and MPEG-4, Dirac video compression standarts due to usage ofDWT. Experimental evaluations show that imperceptibility and robustness performance ofmethod is good. For this purpose, we use Stirmark 3.1 benchmark tools.Keywords: Copy protection, DWT, DCT, digital watermarking, still image watermarking,video watermarking,Thesis Jury1. Prof. Metin YÜCEL (Superviser) Acceptance Date: 27/03/20062. Prof. Dr. Melih Pazarcı Page Number : 893. Prof. Dr. Fikret GÜRGEN4. Assoc. Prof. Dr. Tülay YILDIRIM5. Asst. Prof Dr. Ünal KÜÇÜK

Author

Dr. Serkan Emek

How to Cite

Serkan Emek (Doctorate thesis). Discrete wavelet transform discrete cossine transform based digital watermarking scheme for still images and video signals, 2006, Yıldız Technical University.

Keywords

License

Tüm Hakları Saklıdır

This work is shared under the specified license terms.

More theses from Yıldız Technical University