The Research On Multiple Text Digital Watermark Technology

The technology of the text digital watermark is a cross marginal subject which relates to researches of cryptography, graphics, science of vision, Chinese information processing, image processing, communications, and information security, etc and is a very new research direction in theory at present. So far the acquired achievements on text digital watermarking are mainly focused on the algorithms based on text format. The most typical representatives are algorithms on word-shifting, line-shifting and character-coding. As the watermark information of this kind of algorithm is not inserted into the text content, the algorithms can not resist some common document operations such as format conversion, print processing. Although the text watermarking algorithm based on natural language processing can insert watermarks into text content, computer technology about natural language processing is a bottleneck. When there are several copyrights in a document, the real copyrights are hard to define. Some papers present many protocols to decide which copyright is true. But these protocols are either too complex or don't work well.This paper has given a study of a kind of multiple watermark technology which can not only insert robust watermark information about copyright and load-carrying watermark information about labels but also stamp time in documents. The multiple text digital watermark technology includes three algorithms.(1) The algorithm based on text content. Firstly it pretreats the document and groups the words which satisfy some Chinese grammatical rules according to the rules. Then it inserts watermark information in each word group by transforming words. The document after being inserted watermark still reads fluently and keeps its original meaning. What's more, the inserted watermark always exists in the document if the text content is not changed. This is a good advantage compared with algorithms based on text format.(2) The algorithm based on text format. It extracts the document's Chinese Pinyin, makes planar document three-dimensional by Chinese Pinyin, inserts the watermark information into the solid document by characters'extending, and extracts the watermarks of each layer which verify each other in the end. The algorithm has great payload capability and can be used to carry the information of labels. Through the distilled information we can acquire destructive degree if documents are modified.(3) The zero-watermark algorithm based on time-stamp. It introduces a time stamp organization. Before publishing their works, authors apply to the organization for the time- tamp with combined watermarks indicating author's identity and his works. After successful...