Research On Video Watermarking Algorithms Based On Contourlet Domain

With the rapid advancements of Internet services and various storage technologies, high quality digital data can be copied any times without any fidelity loss which tempt users to spread their own duplicated copies on-line easily and unfettered, which also led to the demand for copyright protection of digital video technology has become more urgent. Digital video watermarking technology has excellent characteristics and a significant advantage in terms of copyright protection which help it become the focus of common attention in the academic and business circles in recent years. In this paper, two kinds of video watermarking algorithms based on Contourlet transform, a new method which has better characteristics compared to Wavelet transform, are proposed. The proposed algorithms have better practicability, and enrich the theory of the video watermarking.The objects of this dissertation are video sequence of the original domain, which the main points are as follows:(1) The research status and significance of digital video watermarking technology are introduced. the basic model, performance requirements, application, classification, a variety of attacks, key technologies and evaluation criteria are elaborated in detail.(2) Starting from structure principle of the Contourlet transform, the basic steps of Contourlet transform are described. The advantages of Contourlet transform in the watermark embedding compared to Wavelet transform are explained.(3) For the weak capacity of most video watermarking algorithms against multiple attacks, a video watermarking algorithm based on pseudo3D-DCT in Contourlet domain is proposed. In the algorithm, the respective advantages of Contourlet transform and pseudo3D-DCT are integrated, which can greatly improve the robustness of the embedded watermark. The experimental results show that the proposed algorithm is robust to MPEG compression and other common video processing on the premise of invisibility.(4) Aimed at the problem of many of existing video watermarking schemes which are not strong robust against specifically attacks (a variety of frame attacks), A video watermarking algorithm based on side view in Contourlet domain is proposed. The main feature of the algorithm is able to resist the unique form of attacks for video watermarking. The experimental results show that the algorithm not only has good visibility but also has strong robustness for noise attack, filtering attack, a variety of frame attacks, MPEG compression and so on.