Multi-class steganalysis for Jpeg stego algorithms

This paper explores two multiclass steganalysis schemes to recognize stego algorithms in use. First of all, Xuan's universal steganalysis is improved to distinguish cover and stego images by the means of applying Bhattacharyya distance to select the most important features. Then, more attentions are paid to design two schemes to recognize stego algorithms with respect to accuracy, reliability and the decision-making cost. Experimental works show that the proposed schemes have satisfactory performance on Jpeg steganography like Jsteg, F5, Outguess and MB2.