Jsteg and F5 are two typical steganography methods of JPEG images and have been used widely. To distinguish F5 stego images and Jsteg stego images, a classification algorithm based on sensitive features and SVM classifier is presented, where the sensitive features are extracted from the subband coefficients of those stego-images and the subband coefficients are obtained by wavelet packet decomposition. Experimental results show that the proposed method can distinguish the Jsteg stego-images from F5 stego-images reliably.
展开▼