| With the development of the cloud computing, more and more businesses orindividuals put their digital products into the "cloud". In order to protect the securityand copyright of the videos in the cloud, this thesis presents a video fractalwatermarking algorithm for cloud computing based on the existing researchfoundation.In this thesis, the algorithm uses a combination of mathematical physics ideas,based on the introduction of the controlling factors as the variance, texture and fractaldimension, improved the existing search algorithm of the fractal coding in the videoframe I, and improved the efficiency of fractal coding. When embedding the robustwatermarking, in the first using the method of the chaotic sequence map to determinethe frame I, then with the difference and threshold of the fractal dimension, in thecollection of the fractal code block of the video frame I, select the strongerself-similarity feature block as the watermark embedding block, and In the frequencydomain, with the human visual system JND model embedded the watermarking in it,in order to certify the ownership and the copyright of the video.In the embedding process of the fragile watermarking, in the first，uses theimproved fractal video coding method to code the video frame I and generate thefragile watermarking signal, then determines the motion vector of embeddingwatermarking in the video frame P and frame B using a threshold value. Finally,achieve the embedding of the video watermarking, with the modifying the greatestcomponent parity of the embedding vector. In this thesis, the extraction process of the two watermarking are both blind.The watermarking algorithm proposed in this thesis, combines thefractal technology and the video watermarking technology, expands the research andthe application area of video watermarking and promotes the development of cloudcomputing in a healthier and faster way. |
