Using Expanded Markov Process and Joint Distribution Features for JPEG Steganalysis

In this paper, we propose a scheme for detecting the information-hiding in multi-class JPEG images by combining expanded Markov process and joint distribution features. First, the features of the condition and joint distributions in the transform domains are extracted (including the Discrete Cosine Transform or DCT, the Discrete Wavelet Transform or DWT); next, the same features from the calibrated version of the testing images are extracted. A Support Vector Machine (SVM) is applied to the differences of the features extracted from the testing image and from the calibrated version. Experimental results show that this approach delivers good performance in identifying several hiding systems in JPEG images.

[1]  William A. Pearlman,et al.  Kernel Fisher discriminant for steganalysis of JPEG hiding methods , 2004, IS&T/SPIE Electronic Imaging.

[2]  Tomás Pevný,et al.  Merging Markov and DCT features for multi-class JPEG steganalysis , 2007, Electronic Imaging.

[3]  Andreas Westfeld,et al.  F5—A Steganographic Algorithm High Capacity Despite Better Steganalysis , 2001 .

[4]  Lisa M. Marvel,et al.  Spread spectrum image steganography , 1999, IEEE Trans. Image Process..

[5]  Yun Q. Shi,et al.  A Markov Process Based Approach to Effective Attacking JPEG Steganography , 2006, Information Hiding.

[6]  Petra Mutzel,et al.  A Graph-Theoretic Approach to Steganography , 2005, Communications and Multimedia Security.

[7]  William A. Pearlman,et al.  Steganalysis of additive-noise modelable information hiding , 2003, IS&T/SPIE Electronic Imaging.

[8]  Alberto Leon-Garcia,et al.  Estimation of shape parameter for generalized Gaussian distributions in subband decompositions of video , 1995, IEEE Trans. Circuits Syst. Video Technol..

[9]  Qingzhong Liu,et al.  Image complexity and feature mining for steganalysis of least significant bit matching steganography , 2008, Inf. Sci..

[10]  Qingzhong Liu,et al.  Feature mining and pattern classification for steganalysis of LSB matching steganography in grayscale images , 2008, Pattern Recognit..

[11]  Jessica J. Fridrich,et al.  Steganalysis of JPEG Images: Breaking the F5 Algorithm , 2002, Information Hiding.

[12]  Siwei Lyu,et al.  How realistic is photorealistic , 2005 .

[13]  Qingzhong Liu,et al.  Image Complexity and Feature Extraction for Steganalysis of LSB Matching Steganography , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[14]  O. Roeva,et al.  Information Hiding: Techniques for Steganography and Digital Watermarking , 2000 .

[15]  Jessica J. Fridrich,et al.  Feature-Based Steganalysis for JPEG Images and Its Implications for Future Design of Steganographic Schemes , 2004, Information Hiding.

[16]  Phil Sallee,et al.  Model-Based Steganography , 2003, IWDW.

[17]  Jason Weston,et al.  Gene Selection for Cancer Classification using Support Vector Machines , 2002, Machine Learning.

[18]  Andrew D. Ker.  Improved Detection of LSB Steganography in Grayscale Images , 2004, Information Hiding.

[19]  Qingzhong Liu,et al.  Feature Mining and Neuro-Fuzzy Inference System for Steganalysis of LSB Matching Stegangoraphy in Grayscale Images , 2007, IJCAI.