A Study On The Technology Of Digital Watermarking In Image And Video

With the rapid development of multimedia technology and the widespread popularity of digital devices, the security of digital multimedia content and the protection of digital copyright have become the focus of attention of the public. Digital watermarking technology realizes copyright protection by hiding the important information in the carriers. Reversible digital image watermarking technology receives extensive attention as lossless digital watermarking technology. Meanwhile, with the rapid development of digital video technology, researching on digital video watermarking technology has become a hotspot. Digital video can be seen as static digital image sequences, so the design and application of digital video watermarking are similar to that of digital image watermarking.The main research of this paper is about reversible digital image watermarking technology and digital video watermarking technology. Firstly, several classic reversible digital images watermarking algorithms are introduced in this paper. And then related video coding standard and several digital video watermarking algorithms are analyzed to help reader to deeply understand relevant technologies. Finally, the following solutions are proposed to protect the copyright of the medical tongue map and digital video.1. Propose a reversible digital image watermarking algorithm for the copyright protection of tongue color image. Though using the strong correlation of original pixel and predicted value, the algorithm predicts the4pixel neighborhood of tongue map, and then embeds watermark by using difference expansion method. Compared to the normal images, the pixels of tongue images are more relevant, so the forecast of tongue image is more accurate, and the images are more likely to embed more watermark information. Experiments show that the algorithm is easy to achieve, the watermark can be fully articulated, the original tongue map vector can be fully recovered, and capacity of watermark is significantly better than traditional method.2. Propose a video watermarking algorithm that bases on corner points and singular value decomposition. This algorithm is frequency-domain algorithm, which mainly uses the characteristics of wavelet theory, corner points and singular value decomposition. Firstly, the algorithm decomposes the brightness channel of digital video image. Then it detects the Harris corner of low-frequency component parts of the wavelet. After K-means clustering the detected points, the algorithm embedded the watermarking with the size of8X8region centering the corner points. Experiments show that the algorithm has good robustness and good watermarked video quality, which demonstrates its good application value.