Research On Text Watermarking Based On Statistical Characteristics Of Inter-Word Characters

The rapid development of computer and Internet technology has led to digital products being distributed deeply and widely to a previously unpredicted level. Meanwhile, their copyright protection is a pressing problem. Fortunately Digital Watermarking has become an effective method of copyright protection for digital products in the last ten years. It has solved the problem of traditional cryptography that encrypted contents are not secure after decryption, therefore it has been studied and used widely as an effective method of knowledge copyright protection and preventing from sham for digital multimedia.There are now many published digital watermarking research papers covering image, video and audio documents, which demonstrate how effective these methods are. However these methods can't be applied to the regular structure of text documents very much. Until now digital watermarking researches on text, have been focusing on the format of text documents because of their particularity. In detail these methods convert binary information into tiny changes in the format of text documents. These algorithms are vulnerable to attack and poor in robustness because they depend totally on the format of text documents, such as line-shift, word-shift and feature coding. Chiefly this is because the watermark data can only be embedded onto the outside of the text document's content but can't be embedded into the inside of it. Therefore to solve this problem we will have to study text-watermarking algorithms based on the content of text documents.This paper not only uses the idea of text watermarking based on format coding, but also attempts to study the content and the format of inter-word characters in English text documents. So this text-watermarking algorithm is based on the content and the format of text documents, that is text watermarking based on statistical characteristics of inter-word characters is presented. This method takes out the special characters i.e. inter-word characters from the English text documents, studies their content and format and uses them as the foundation for loading the watermark data. In addition this method doesn't just use a single inter-word character as the carrier for loading a watermark bit. Here all the words in a text document are classified depending on specific features. Next adjacent words making up a sentence are grouped, then this group is also classified. Finally the statistical characteristics of inter-word characters in the same class of groups throughout the text document are calculated and these become the basis for loading the...