A Statistical Blind Image Steganalysis Based on Image Multi-classification

In this paper, we proposed a new statistical framework for blind image steganalysis that is shown to be of higher detection performance accuracy than truly current steganalysis systems. Therefore, we have introduced a multi-classification methodology based on image features to group images into the optimal classes in order, to make the models specific and differentiated all the infected images. To distinct images more effectively and thus improving the system accuracy, we have applied Gaussian Mixture Model (GMM) and also an unsupervised algorithm to learn a finite mixture model. Afterward, for each images class we can select and design a specific model of steganalyzer. We have also employed Support Vector Machines (SVMs) to design the models of steganalyzer. Proposed framework can enable us to employ multivariate features extracted from different domains in order, to obtain much better distinction of images and also better designing the steganalyzers in which, can be applied to any types of steganalysis techniques. The result comparison shows the advantages of the proposed framework over the current and prior steganalysis systems with the best overall results.