Text Digital Watermarking Algorithm Based On Font Style Distinction Of Chinese Character

In recent years, with the emergence and continuous development of the Internet, the way of people exchange information has changed a lot. The traditional means of exchange information using paper, tapes, videos and other physical carrier has been gradually replaced by the new means of exchange information using documents, audio files, video files and other binary files carrier base on information technology. Now, the literary works written by writers, the production of teaching resources created by the teachers or experts in one field, the music products and various television programs and movies published and distributed by music producers and film makers have been put on the Internet and exchanges. This situation has brought two effects: on the one hand the emergence of the Internet makes information more effectively to be passed and distributed, makes it possible to get useful information in a little time; But on the other hand this mean of exchange information will lead to some uses who wasn't authorized get the interests which should only authorized users can have. The end, this situation harms the interests of the creators of the information resources, greatly damaged the creative passion of creators of information resources. Thus, it's urgent that people want to get a technical, which can continue to play a positive aspect of the Internet, at the same time, it can be used to protect the various information resources that have copyright. The digital watermarking technology generates and develops against this backdrop.Currently, digital watermarking technology has been widely applied to images, audio, video, documents, software, database and other application areas, and has achieved good results of the application; thereby gradually get further people's attention. But at the moment there is a situation that research results which based on image, audio and video digital watermarking is the majority, and many of them are very effective, but these methods are not applicable to the rules of text documents. So far, the text-based digital watermarking study due to the special nature, relevant research results mainly concentrated in text-based format. Through make some adjustments on the format of the document, hide the watermarking information into text documents with binary code by the document object changes. As such algorithms rely entirely on the text format (line shift, the word shift, feature coding, etc.), there are the shortcomings of not strong anti-offensive, the poor robustness. The reason for this is mainly because the watermark can only join out of the text, but unable to join the contents of the text. Therefore, it must be that only we study the text content-based digital watermarking algorithm, we can solve this problem.This paper studies the relationship between multi-graphs and the human cognitive habits of the Chinese characters, and split one character into a number of components, then Each component as a graphic unit, and according as the relationship of multi-graphs, recombine these components using different way. Then one Chinese character will have multi font style. Then, through using different font style, embed the water mark into the text. This algorithm embeds the watermark information into the contents of the text successfully. Compared with the previous watermarking algorithm, this algorithm gets a big step forward on aspect of invisible character and robust character. Finally, this paper approves that this new algorithm can be applied to electronic documents and paper documents both by the theoretical analysis and experimental analysis. And what's more, it also approves that this new algorithm has strong invisible character and robust character.