Research On Information Hiding And Detection Of Hidden Information For Webpage

In the era of the Internet, the conflict of information hiding and detection of hidden information has become a main topic in information warfare. The study of this conflict is of great significance for protecting the national security and social stability. On the one hand, the covert communications become safer with information hiding. On the other hand, with the rapid increase of information on the Internet, the abuse of information hiding may be a big threatens to national security and social stability. It is an urgent task to bring the study of information hiding and detection of hidden information into the framework of information security to confront the increasingly fierce and complex information warfare.Static webpage (webpage for short) is the main carrier of information released on the Internet. The development of the Internet is featured as open, global, dynamic, shared and seamless connected. Personal privacy and secret messages are easily revealed, and illegal messages can be very easily transmitted within webpage .Research on webpage information hiding for protecting personal privacy and secret message from revealing and detection of hidden information in webpage to monitor illegal message transmission has a great prosperity for future applications.This dissertation mainly focuses on information hiding and detection of hidden information algorithms of webpage, and stego-webpages searching on the Internet.The main contributions are summarized as follows.(1) Based on the scheme of importing the Class Selectors in CSS repeatedly, a novel webpage information hiding algorithm is proposed in this dissertation. Class Selectors of some objects in the Operable Block of CSS are imported repeatedly to embed information. Because of the close integration of the place information embedded and the content of the webpage, it can enhance the ability of the algorithm to contradict with detection and auto filtration. Experimental result shows that this algorithm has preferable imperceptibility and a large capacity of information embedding.(2) Based on equal tag, an algorithm of webpage information hiding is presented. This algorithm overcomes the shortcoming of poor imperceptibility of traditional algorithms, and has a larger capacity than that of the algorithm based on attributes pair order. According to the embedded rule, firstly the message M is changed into a large integer N , and then embedded by changing the original tag to its equal tag. The algorithm doesn't lengthen the cover-webpage, and has good imperceptibility.(3) In order to resolve the conflict between imperceptibility and embedded capacity in information hiding, a multi-webpage embedding technique is presented. Sequence and average embedding schemes are designed to ensure preferable imperceptibly after a high volume message is embedded. A system of webpage information hiding is implemented, which integrates four algorithms and has multi-webpage embedding function.(4) An algorithm of detection of hidden information is presented to detect invisible characters hidden in webpage. This detection algorithm can estimate the length of secret messages, and has a nice detection rate and a lower false positive rate.(5) After deep research on webpage information hiding algorithms based on the uppercase-lowercase conversion of the tag letters, the singularity of tag mismatch is found when the secret messages is embedded. For this reason, an algorithm of detection of hidden information in webpage is presented. According to the tag mismatch, the embedded capacity rate is calculated to indicate whether the webpage has secret messages. This algorithm can be simply implemented, and can estimate the length of the secret messages of the stego-webpage.(6) Secret messages can be embedded into letters in tags of a webpage, which are imperceptible to human eyes if viewed with a browser. These messages, however, alter the inherent characteristic of the correlation of letters in a tag. Based on the correlation of letters, a new detection algorithm is presented for detection of hidden information in webpage. The experimental results show that this algorithm has a nice detection rate and a lower false positive rate.(7) Based on web crawler, an active search system"WebDetector"is implemented to search the stego-webpages on the Internet. The detection algorithms in chapter 5 and chapter 6 are integrated in the system. The architecture of passive search based on network traffic analysis is designed, and the function of each module is presented. The advantages of active and passive search are combined to set up a general search framework to search the stego-webpages on the Internet.