Research On Zero-Watermarking Algorithms For Chinese Prose Writings

With the continuous and deepening application of Internet,a large number of digital works spread through the Internet.At the same time,it has gradually become a trend for literary works to be published on the Internet.However,due to the openness and reproduction of network communication,the copyright protection of these digital literary works needs to be solved urgently and text digital watermarking technology is one of the effective means to solve this problem.On the basis of analyzing the existing text watermarking schemes and aiming at the characteristics of prose writings,this paper designs a zero-watermarking scheme for chinese prose writings and discusses the scheme in detail.The specific research work has the following aspects:(1)Based on the analysis and comparison of existing text digital watermarking schemes,a zero watermarking scheme for chinese prose writings was designed according to the characteristics of prose writings.The scheme includes three parts: text preprocessing,watermark construction and watermark detection.(2)In the part of text preprocessing,the specific workflow and model for text preprocessing are given.And the key modules such as word segmentation,POS tagging,screening for post-segmentation words,word frequency statistic and word weight calculation are also designed and implemented.(3)In the part of watermark construction,based on the semantic and statistical characteristics of prose writings,this paper gives the work flow and the model of the construction process.And the key modules such as the key word set extraction,the core verb set extraction,the proportional feature of adjective extraction,the feature coding,the feature fusion and encryption and the watermark generation are designed and implemented.(4)In the part of watermark detection,the copyright ownership is determined based on similarity calculation,the specific workflow and the model of watermark detection are given.The algorithms of the key modules such as the key word set similarity calculation,the adjective proportional feature similarity calculation and the copyright determination are designed and implemented.(5)Experimental analysis and comparison of the proposed watermarking scheme are carried out from the aspects of robustness,error rate and security.The results show that the scheme designed in this paper has strong pertinence and applicability for the copyright protection of prose writings.