Investigating and analyzing steganalysis of hidden data using highly undetectable steganography method

Steganalysis is divided into two main stages of extracting features from the information studied for Steganalysis and classifying the features extracted from them. In this article, theoretical suggestions for designing the Steganalysis algorithm with optimal accuracy and speed are presented. More specifically, two basic proposals are presented in this section. The first suggestion is to use the thin segmentation dictionary model. This model, in addition to the advantages of generally using dictionaries for display retrieval, i.e. reducing the dimension of the thin domain and as a result a significant reduction of the computational load, with the discriminating feature that is added to it, increases the accuracy of the system. The second suggestion is to use the double tank display, which will be reviewed at the end of the article. The use of the presented solution causes the dimensions of the optimization problem that is used to obtain the thin display to be significantly reduced, and as a result, the computational load is reduced and the speed of the system is increased. As an objective example, Highly undetectable steganography steganography method is investigated using this method and traditional methods.

Keywords:

steganalysis. steganography , highly undetectable steganography