Normalized Watermarking Algorithm Based On The Scrambling Processing And Geometric

Digital watermarking is a new technique for information security. It embeds a special invisible mark into digital multimedia data, men carries out the functions such as information hiding, storing, copyright, protection by tracking the use of the image by detecting the watermark. At present, watermarking technique has been a kind of effective means to multimedia copyright authentication and its integrity protection.This paper takes still images as example.Contrapose at the frailness of current algorithms, this paper give a algorithms based on the technique of image scramble and geometric normalization. At first we give a brief introduction about image scramble, and then analyse how to value the scramble degree. Then we discuss the geometric normalization and give my normalization method.At last we give a blind watermarking algorithm based on wavelet. After many experiments, we conclude that the arithmetic express well both in imperceptibility and robustness.This paper will be divided into six parts. First part introduce the status of current study,the background of this paper and the basic theory in the field of the digital watermarking. Then introduce the theory of the image scramble. Then experiment to confirm the best scrable degree, and conclude that the method based on the part's characteristic is better than the method based on the change of the location. The third part introduce the technique of geometry normalization. Firstly this paper analyse one of the normalization methods based on the geometry normalization. Then raise a DIY way to achieve geometry normalization and go with the detailed mathematic testify. The fourth part give the algorithm of watermarking, Firstly do geometry normalization to the host image.Then get the most important part based on the theory of changeless of object's mass center.and then we embed the watermarking into the DWT coefficient of the important part. We need not the quondam image or the watermarking image when extracting, so the algorithm is one of the blind watermarking algorithms. In The fifth part, I do lots of experiments to validate the capability of the algorithm when it's under attack.As result we find the algorithm is robust, especially in the shearing and affine attack.At last we do the compare between this algorithm and two others. All the experiments prove that the algorithm do well in both imperceptibility and robustness.Finally,it's the overall summarization of the paper and research expectations of the digital watermarking.