Feature Selection for High Dimensional Steganalysis

摘要

In today's digital image steganalysis, the dimensionality of the feature vector is relatively high. This may result in much redundancy and high computational complexity. In this paper, a novel feature selection method is proposed from a new perspective. The main idea of our proposed feature selection method is that the element in the extracted feature vector should consistently increase or decrease with the increase of embedding rate for a given steganographic scheme. Various experimental results tested on 10000 grayscale images demonstrate that our feature selection method can reduce the dimensionality of the high dimensional feature vector efficiently, and meanwhile the detection accuracy can be well preserved.