Text Zero-Watermark Based On Topic Extraction

With the widespread use of Internet and other communication technologies in recent years, the network delivery of novels, journals and documents in digital way is more and more popular. So how to effectively realization of the electronic document copyright protection and prevent the diffusion of the unauthorized document’s contents has become one of the key issues in the field of document security. At present, the technology of text digital watermarking is considered an effective method to protect the copyright of digital produces.Based on the analysis of the existing text digital watermarking methods, a new document watermarking schema is proposed in this paper. The main research work is summarized as follows:(1) Firstly, in theory, summarized the existing digital watermark and text digital watermarking technology program, and analysed the strengths and weaknesses of the existing watermaking algorithm, moreover, summarized the general principles and methods of the text zero-watermarking scheme.(2) Based on topic segmentation and space model techniques constructed a text zero-watermark scheme, planning and design of the overall scheme model and the process of watermarking, and then designed a structure model and a detection model of the zero-watermark.(3) In the process of constructing the text zero-watermark, it introduce the concept of semantic distance to divide the topic paragraph, and use to saved the topic sentences by space model. Meanwhile, we discuss the main module of watermark construction in detail, which include keywords extraction, topic segmentation, topic sentence extraction,3D model construction, feature extraction, and zero-watermark generation etc, and present the detailed algorithms for them.(4) In the process of detecting the text zero-watermark, according to reconstruction of the text zero-watermark and archive watermark information, take advantage of semantic similiarity calculation to determine the ownership of copyright. Meanwhile the key algorithms of words and sentences semantic similarity calculation.are designed.(5) Finally, the choice of parameters, accuracy of summary extraction and anti-attack ability of the proposed algorithm were proved with simulation experiments. The result of the simulation shows that our summary extraction algorithm is more accurate and efficient than traditional methods, what’s more, the text zareo-wartermark has good robustness, security and anti-attack ability.