Research Of Digital Watermarking Techniques Based On Image And Video

With the rapid development of the digital multimedia technology, it is being easier to store,reproduct and distribute the mutimedia information, and it has improved people's material and cultural life. It is convenient to access to the mutimedia data through the Internet and CD-ROM,whereas it provides the chance to infringe the copyright such as replicate,unauthorized access,tamper and distribution for various illegal purposes, and the copy of each on the analog information will have different degrees of distortion. So how to ensure the information security and effectively protect the legitimate interests of copyright owners is becoming an urgent issue. So the technique of digital watermark is being developed under this circumstance and has been the main method of multimedia copyright protection. It is to embed the copyright information, according to some particular algorithm, into digital original signal, such as image, audio, video, and the like, to realize the copyright protection and authentication. And it has been the focus in the field of multimedia signal processing as an important branch of the information security technology. The thesis of the image and video digital watermarking technology has been studied and several specific watermarking scheme has been proposed, including the following:First of all, the principles of the digital watermarking technology,framework,characteristics,classification and attack methods were summarized and explained, and in this thesis the current general performance of the watermarking algorithm evaluation criteria were introduced and analyzed. It laid a foundation for the specific algorithm in the next step.Secondly, the thesis proposed a DWT-based image watermarking algorithm through introducing the basic principles of wavelet transforms, definition and the analysis of application,especially muti-resolution wavelet analysis. The binary image was chosen as the watermark. First, the original image was transformed through the 2 Haar wavelet, and we have the high frequence subband coefficients of LH2,HH2 and HL2, by three points, the same position coefficient value compared to embed watermark. We use MATLAB to simulation the algorithm, and it can maintain the image quality, but also the common compression,shear,noise,rotation and other attacks have good resistance to attack, the algorithm has achieved blind watermark extraction.And then, the RGB color video watermarking algorithm is proposed based on independent feature of consecutive frames, the embedding scheme is extended from the image to the field of video, while extraction by principal component analysis and independent component analysis of 2 methods, in comparing the 2 methods by extracting the feature from consecutive frames of the RGB video image, ICA analysis is better than PCA. Then embedd watermark into the static component or the movement component, these two programs by comparing the simulation experiments, we can see that they have different performances and robustness to the different attack, but both attacks for anti-MPEG is to be further study.As a new technology, digital watermarking has a rapid development in the last few years, and more and more scholars began to concern about the areas of research, but there are still many problems needing urgent solutions and does not touch on some topic in the development. It is necessary to form a set of theory,the standard of the watermarking technology,performance evaluation of watermarking system and the application of system. Finally, the paper summarizes the digital watermarking technology and the development of this field were prospected.