The Research On Robust Text Digital Watermarking Based On Structure Knowledge Of Chinese Characters

With the popularization of Internet and the development at full speed of the multi-media technology, the copyright protection of digital works has already become the hot issue at present. Many watermarking algorithms have been proposed to protect copyright in an age where digital works may be easily and illegally modified or reproduced. Since a lot of important files and other valuable information stored as text style are transmitted on the network, people can use multimedia treatment software to edit and modify text data conveniently if no protective measures for texts are taken. It is obvious that the research on the copyright protection of texts is no time to delay. Text digital watermarking technology offers an effective scheme to solve above problems by embedding given watermarking information in carrier and extracting it from watermarked works. At present, the most research on robustness of text watermark is focusing on embedding strategy, but the robustness of text watermark is determined mostly by two aspects, namely watermarking structure and embedding strategy.By researching on character coding and the Chinese mathematical expression theory, a novel robust text watermarking algorithm based on structure knowledge of Chinese characters is proposed to solve the problem of copyright protection for Chinese texts in this paper, which watermarking structure and embedding strategy are comprehensively considered. The Chinese mathematical expression theory can remedy the inadequacy of information of a Chinese character ISN(internal statement number), which ensuring that a computer handles Chinese characters in Chinese components whose granularity is much smaller than Chinese characters, which creates advantages for the research of text watermark. This method obtains structure types and strokes of Chinese characters through Chinese characters mathematic expression, and then the whole document is divided into two blocks by utilizing structure types of Chinese characters. In each block, the position selected to embed is chose by Chinese character's strokes and the watermarking bit. Watermarking bits with a robust structure is generated by chaos encrypting and hamming checkout coding; embedding is implemented through setting underline to font. When the watermarking bit is destroyed, which position can be judged exactly, and then block checkout and hamming checkout are used to recover it. Meanwhile, watermarking can be extracted without the original document and original watermarking information, which is significative copyright information, so it is a public watermarking technology. The experimental result indicates that the algorithm has preferable transparency and robustness.