Blind image steganalysis via joint co-occurrence matrix and statistical moments of contourlet transform

Abstract

A blind color image steganalyzer is proposed, in which the features are extracted from Contourlet domain. Statistical features of Contourlet coefficients and cooccurrence metrics of subband images are used as features. For evaluating the proposed steganalysis method, some popular steganography methods such as OutGuess, JPHS, Model-based and Jsteg are used with payloads of 10% to 25%. To reduce the number of features, Analysis of Variance (ANOVA) method is used and the selected features are fed to nonlinear Support Vector Machine (SVM) for classification into stego and clean images. Empirical results show high sensitivity of Contourlet and co-occurrence matrix features to data hiding.