Steganalysis using features based on Markov Mesh Models
Xiayang Shi, Bei-bei Liu, Yongjian Hu · 2014
Although numerous steganalyzers for least significant bit (LSB) matching have been presented, the detection for uncompressed images and low embedding rates remains challenge for steganalysts. In this paper, we propose a novel method for detection of LSB matching steganography, which is based on the features extracted from a conditional probability matrix described by Markov Mesh Models (MMMs). The extracted features are calibrated in image domain by image calibration technique to improve the detection rate. Support vector machine (SVM) is employed to classify the images with/without hidden message. Extensive experiments show that the proposed scheme outperforms the state-of-arts LSB matching steganalysis methods.