Selection of image features for steganalysis based on the Fisher criterion |
| |
| Institution: | 1. Zhengzhou Information Science and Technology Institute, Zhengzhou, Henan 450001, China;2. State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan 450001, China;3. State Key Laboratory of Information Security, Institute of Information Engineering, Chinese Academy of Sciences, Beijing 100093, China;1. Inserm U1016, Institut Cochin, Paris, France;2. CNRS UMR8104, Paris, France;3. Univ Paris Descartes, Paris, France;4. Inserm CIC BT505, CIC de Vaccinologie Cochin Pasteur, Paris, France;5. Assistance Publique-Hôpitaux de Paris (AP-HP), Hôpital Cochin, Paris, France;6. Institut National de Santé et de Recherche Médicale, INSERM U976, Saint-Louis Hospital, Skin Research Center, 75010 Paris, France;7. Paris Diderot University, Sorbonne Paris Cité, Laboratory of Immunology, Dermatology & Oncology, UMR-S 976, 75010 Paris, France;1. ICD - LM2S - UMR STMR CNRS - Troyes University of Technology, 10004 Troyes Cedex, France;2. ICD - LM2S - UMR STMR CNRS - EPF École d׳ingénieurs, 10000 Troyes, France;3. I3S - University of Nice Sophia-Antipolis - UMR7271 - UNS CNRS, 06900 Sophia Antipolis, France |
| |
| Abstract: | A steganalytic feature selection method based on the Fisher criterion used in pattern recognition is proposed in this paper in order to reduce effectively the high dimensionality of the statistical features used in state-of-the-art steganalysis. First, the separability of each single-dimension feature in the feature space is evaluated using the Fisher criterion, and these features are reordered in descending order of separability. Then, starting from the first dimension of the reordered features, as the dimension increases, the separability of each feature component is analyzed using the Fisher criterion combined with the Euclidean distance. Finally, the feature components with the best separability are selected as the final steganalytic features. Experimental results based on the selection of SPAM (Subtractive Pixel Adjacency Matrix) features in spatial-domain steganalysis and CC-PEV (Cartesian Calibrated feature extracted by PEVný) features in DCT-domain steganalysis show that the proposed method can not only reduce the dimensionality of the features efficiently while maintaining the accuracy of the steganalysis, but also greatly improve the detection efficiency. |
| |
| Keywords: | Image steganography Steganalysis High-dimensional feature Feature selection Fisher criterion |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
