Study On Image Watermarking Scheme Robust Against Geometric Attacks

With the rapid development of multimedia and Internet techniques, the multimedia security becomes an important and challenging research topic. Digital watermarking technology is an effective method for digital content protection such as copyright protection and becomes an important research area in the field of information hiding. Thus the research on digital watermarking is of great importance. At present, the ability of counteracting geometric attacks hasn't completely achieved the request of application. The geometric attack is considered as a pivotal problem to improve the robustness of watermarking algorithm. Considering the problem in information security and copy protection, this dissertation makes a wide and deep research on the current hotspot in digital watermarking technology, such as the image watermarking technology against geometric attacks.The main research contents of this dissertation are recapitulated as follows:1. A new robust digital image watermarking based on pseudo-Zernike moments and image normalization technique is proposed. Firstly, the geometrically invariant space is constructed by using image normalization. Then, the pseudo-Zernike moments of the normalized image are computed. Finally, some low-order pseudo-Zernike moments are selected, and the digital watermark are embedded into the host image by quantizing the magnitudes of the selected pseudo-Zernike moments. Experimental results show that the proposed scheme is not only invisible and robust against common signals processing such as filtering, noise adding, and JPEG compression, but also robust against the geometric attacks such as rotation, scaling, translation and combination attacks.2. Based on multiscale theory and image normalization technique, a new feature-based image watermarking scheme robust to geometric attacks is proposed in this paper. Firstly, the Harris-Laplace detector is utilized to extract steady feature points from the host image. Then, the local feature regions (LFR) are ascertained adaptively according to the feature scale theory, and they are normalized by image normalization technique. Finally, according to the predistortion compensation theory, several copies of the digital watermark are embedded into the nonoverlapped normalized LFR by comparing the DFT (Discrete Fourier Transform) mid-frequency magnitudes. Experimental results show that the proposed scheme is not only invisible and robust against common signals processing such as median filtering, sharpening, noise adding, and JPEG compression etc, but also robust against the geometric attacks such as rotation, translation, scaling, row or column removal, shearing, local geometric distortion and combination attacks etc.