Font Size: a A A

The Research On Text Information Hiding Based On OOXML Format

Posted on:2016-01-15Degree:MasterType:Thesis
Country:ChinaCandidate:C X SunFull Text:PDF
GTID:2348330488482026Subject:Engineering
Abstract/Summary:PDF Full Text Request
The text is one of the most frequently and widely used digital carriers in daily life and work. The digitization and networking of text information provides a great of convenience for people. Meanwhile, it brings a series of security problems, such as the loss of important information the privacy disclosure the plagiarism of works, and so on. Fortunately, text information hiding techniques can well solve the security problems above, mainly including two categories: steganography and digital watermarking. This thesis focuses on researching text information hiding related to OOXML format documents. The main contents of this paper are as follow:(1) A steganographic method based on the revision identifiers is proposed, which preserving the features of OOXML format documents. The method embeds secret information by replacing part of the values of the revision identifiers, while ensuring the statistical characteristics of the modified values accorded with those of the normal ones. Meanwhile, position marks are added to track the locations of the embedded information. In order to keep the internal data consistent of the document, the new created values are added into other related parts. Experimental results show that the method not only possesses good imperceptibility and anti-detection ability, but also has high security and large embedding capacity.(2) A multiple digital watermarking scheme is proposed, which can be used for copyright protection tamper detection and localization in OOXML format documents. The scheme embeds the robust watermark through synonymy substitution in the robust watermark embedding stage. Then embed two fragile watermarks through the methods based on text split and revision identifiers in the fragile watermark embedding stage. In the extraction process, the robust watermark is extracted by the robust watermarking algorithm to authenticate the copyright holder, and then the two fragile watermarks are extracted from the text content and revision identifiers. At last, the three watermarks mentioned above are compared to judge whether the document has been tampered and where the tampers happen. The experimental results show that the multiple watermarking scheme has a good imperceptibility. It not only performs well on authenticating the copyright of the documents, but also realizes the tamper detection and localization.
Keywords/Search Tags:Information hiding, OOXML format, Steganography, Digital watermarking, Revision identifiers, Synonym substitution, Text split
PDF Full Text Request
Related items