Research On 3D Mesh Model Zero Watermarking

watermarking system. These methods have to change the original mesh data to achieve the watermarking. Because the 3D mesh is more sensitive and complex than 1D (audio) and 2D (images and video) signals, the disturbing free watermarking framework named zero watermarking system should be developed.In order to provide some references for the future work of the watermark, this paper proposes two kinds of zero watermarking frameworks. The parameterization based zero watermarking framework is an expansion and improvement to the 2D watermarking methods. After the planar parameterization, the original 3D geometry signal has been transformed to the 2D domain. Then the zero watermarking vectors can be extracted from the coordinate mapping matrix. Another framework suggests to analysis the 3D mesh data directly. Different from the former framework, the direct analysis frame discard the parameterization method and employ the 3D signal analysis algorithm to process the mesh data. This scheme can not inherit the 2D analysis algorithm, but it keeps 3D features of the model pretty well.Two different implements of the parameterization based framework are introduced at end of the thesis. The spatial domain algorithm applies the SVD to analysis the coordinate mapping matrix. By using FFT, the frequency domain algorithm transforms the mapping matrix to the frequency data and extracts zero watermarking vectors from it. Due to the difference between source signals, the method used to test the similarity is different, but both measurement standards are the same. To the standard attacks, the experiment result shows that the frequency method has better robustness than the spatial one.