A Coutourlet-Based Image Watermarking Using Generalized Gaussian Distribution Model

With the increasing demands of copyright protection, digital watermarking has being paid more and more attention. In the design of a watermarking method, the modelling of signal by a general parametric family of statistical distributions plays an important role in many signal processing applications. Some conventional methods based on Gaussian distribution to model the image coefficients in the transform domain. In this paper, I proposed to adopt the generalized Gaussian distribution (GGD) for modelling the contourlet transform sub-band coefficients and for image watermarking scheme. The contourlet transform (perfect reconstruction and directional selectivity) is considered. Its improved robustness and imperceptibility are due to embedding in the directional subband with the highest energy. In watermark detection, the Neyman-Pearson (NP) detector is used to detect the watermark. Experimental results show that the effectiveness of the presented watermarking method and its robustness against common image processing and some kinds of geometric attacks.