Steganalysis for JPEG images based on weighting fusion and Markov matrix
Zhiping Zhou · Jisuanji yingyong yanjiu · 2009
This paper designed a novel classifier based on feature fusion to attack the advanced JPEG steganographic methods.Applied Markov transition matrix to derive correlations between the coefficients,which was extracted respectively from horizontal,vertical,and zigzag difference arrays in DCT domain to construct local Markov feature.Distributed the feature weight according to the contribution degree on classification,used SVM to classify the generated weighted feature vectors.The experiment results demonstrate the effectiveness of the proposed method,the detection rates of the weighted features(4:3:3) are always higher than 91% while attacking the four steganographic schemes(Outguess,F5,Mb1 and Mb2) at imbedding rate 0.05,and the feature fusion operation does not increase the feature dimension.