Enhancing steganalysis accuracy via tentative filtering of stego-containers

We introduce a new approach to steganalysis called “the tentative filtering” and consisting in inserting an additional filtering phase before the final classification in order to select those containers where stego-information can be reliably detected. The size of this “good” subset of containers can be considered as an additional characteristic of the detector. We introduce three methods for implementing the tentative filtering: the naive method, the simple classification, and the combined classification. The experiments demonstrate that it is possible to select about 35 % of BOSSbase v1.01 images, for which HUGO 0.4 bpp is detected with the error less than 0.003, while the error over the whole set is 0.141. It is also demonstrated that it is possible to select about 5 % images, for which HUGO 0.1 bpp is detected with the error less than 0.05, while the whole set gives the error 0.37 (which is not quite a reliable detection).