Local Correlation Pattern for Image Steganalysis

摘要

Correlation of pixels is the most important information used for image steganalysis. Current methods often consider some special types of relationships among neighboring pixels. In this paper, we propose a general descriptor to consider the correlation of pixels comprehensively. We consider the correlation of pixels in an adjacency pattern as a local correlation pattern (LCP). The LCP descriptor is proposed to embrace different local correlation patterns and represent each pattern by mapping the relative values of pixels in the pattern to a numerical value. Then, histograms of LCP values are taken as features for steganalysis. The LCP descriptor also can be used for describing the correlation of elements in the residual image obtained by image filtering. Experiments show that our constructed feature set based on the LCP descriptor outperforms a state-of-The-Art method on detecting three popular steganographic algorithms.