Image Watermarking Algorithm Research And Realization Matrix Transform

Digital watermarking technology provides an effective way to protect copyright and track down the infringement of copyright, and is known as the last line of defense in multimedia content protection. The digital watermarking technology which uses image as a carrier is a hot research focus in current watermarking technology.Images may be attacked intentionally or unintentionally during the transmission process, therefore, the design of watermarking algorithm requires not only the ability to better resist various attacks (i.e. the robust), but also the capability to easily implement (i.e. less computation). Satisfying these two requirements at the same time is a critical problem when designing image watermarking algorithm.The main content and research work in this thesis include:(1) The basic notion, feature, classification and applied area of digital watermarking are systematically introduced. The attack types of digital watermarking are analyzed and its performance assessment standard is provided.(2) A digital image watermarking algorithm based on discrete cosine transform is systematically studied. Experiments show that the algorithm not only has good concealment, which means the extracted watermark under no attack conditions are exactly the same as the original watermark, but also has comparatively good robustness to noise, filtering, rotation, scaling, and shearing.(3) Based on the above, this paper presents a digital image watermarking algorithm based on matrix transformation. In view that the Householder transformation matrix has the characteristics of symmetry, orthogonality and disturbance, the image is Householder transformed and decomposed as a matrix. Through a series of simulation experiments which attack on the algorithm, and compared with DCT algorithm, the experimental results show that, the algorithm is significantly better than the DCT algorithm in terms of computational complexity and robustness.