Research On Text Information Hiding Based On Lossless Compression

Information hiding technology imperceptibly embeds secret information into the public digital carrier,so as to achieve the purposes of covert communication,copyright protection,and so on.With the rapid development of Microblog,E-commerce,World Wide Web,etc.,the volume of text data is growing rapidly,thus text is a good carrier for information hiding and text-based information hiding technology has attracted more and more scholars' attention.In view of the lacks of embedding capacity and security of text information hiding,this paper improves the performance by using secret information compression and text selection,which can shorten the length of embedded information to increase the embedding capacity,reduce the number of modification operations and reduce the distance between original text and stego text to improve the security of stego text.The main research results are as follows:In order to improve the embedding capacity and anti-steganalysis capability of the linguistic steganography,we propose a method of text information hiding based on LZW compression and synonym substitution.Firstly,the lossless compression algorithm LZW is used compressing the secret information,which can reduce the length of the embedded information.Then,a synonym encoding rule is designed by analyzing the distribution characteristics of the LZW compressed secret information,which can make the relative entropy as small as possible between the digitalized synonyms in the cover and the corresponding stego text,and enhance the anti-steganalysis capability.The experimental results show that the proposed method can effectively improve the embedding capacity,reduce the relative entropy between the cover text and the stego text,and decrease the possibility of being detected by linguistic steganalysis.In order to further improve the embedding capacity and undetectability,this paper proposes an information hiding method based on path compression and text selection.First,we propose a minimum maximum weight algorithm(MMW)to choose a best path for the secret information words appearing in each candidate cover text,which can be employed to reversibly and losslessly compress the retrieved secret words.The unretrieved secret words are compressed by Huffman coding.Then,according to the length of the compressed information,ten candidate cover texts are selected from a large-scale text set and are embedded their compressed secret information by using synonym substitution.The final stego text is chosen from ten candidate stego texts by using a given selection rule derived from the distance between a cover text and its stego text.Experimental results show that the proposed method achieves better compression ratio than Huffman and LZW coding algorithms;compared with the similar methods,it also has highest embedding efficiency,and performs best in security with the text selection rule.