Steganalysis of LSB Replacement Using Parity-Aware Features

Detection of LSB replacement in digital images has received quite a bit of attention in the past ten years. In particular, structural detectors together with variants of Weighted Stego-image (WS) analysis have materialized as the most accurate. In this paper, we show that further surprisingly significant improvement is possible with machine---learning based detectors utilizing co-occurrences of neighboring noise residuals as features. Such features can leverage dependencies among adjacent residual samples in contrast to the WS detector, which implicitly assumes that the residuals are mutually independent. Further improvement is achieved by adapting the features for detection of LSB replacement by making them aware of pixel parity. To this end, we introduce two key novel concepts --- calibration by parity and parity-aware residuals. It is shown that, at least for a known cover source when a binary classifier can be built, its accuracy is markedly better in comparison with the best structural and WS detectors in both uncompressed images and in decompressed JPEGs. This improvement is especially significant for very small change rates. A simple feature selection algorithm is used to obtain interesting insight that reveals potentially novel directions in structural steganalysis.

[1]  Jessica J. Fridrich,et al.  Steganalysis in high dimensions: fusing classifiers built on random subspaces , 2011, Electronic Imaging.

[2]  Nasir D. Memon,et al.  On steganalysis of random LSB embedding in continuous-tone images , 2002, Proceedings. International Conference on Image Processing.

[3]  Jessica Fridrich,et al.  Steganography in Digital Media: References , 2009 .

[4]  Xiangyang Luo,et al.  An Improved Sample Pairs Method for Detection of LSB Embedding , 2004, Information Hiding.

[5]  Sorina Dumitrescu,et al.  Detection of LSB Steganography via Sample Pair Analysis , 2002, Information Hiding.

[6]  Jessica J. Fridrich,et al.  Steganalysis of Content-Adaptive Steganography in Spatial Domain , 2011, Information Hiding.

[7]  Florent Retraint,et al.  A Cover Image Model For Reliable Steganalysis , 2011, Information Hiding.

[8]  Jessica J. Fridrich,et al.  Ensemble Classifiers for Steganalysis of Digital Media , 2012, IEEE Transactions on Information Forensics and Security.

[9]  Andrew D. Ker.  A General Framework for Structural Steganalysis of LSB Replacement , 2005, Information Hiding.

[10]  Sorina Dumitrescu,et al.  Detection of LSB steganography via sample pair analysis , 2002, IEEE Trans. Signal Process..

[11]  Andrew D. Ker.  Optimally weighted least-squares steganalysis , 2007, Electronic Imaging.

[12]  Masoud Nikravesh,et al.  Feature Extraction - Foundations and Applications , 2006, Feature Extraction.

[13]  Jessica J. Fridrich,et al.  On estimation of secret message length in LSB steganography in spatial domain , 2004, IS&T/SPIE Electronic Imaging.

[14]  Jason Weston,et al.  Embedded Methods , 2006, Feature Extraction.

[15]  Wei Su,et al.  Steganalysis based on Markov Model of Thresholded Prediction-Error Image , 2006, 2006 IEEE International Conference on Multimedia and Expo.

[16]  Rainer Böhme,et al.  Revisiting weighted stego-image steganalysis , 2008, Electronic Imaging.

[17]  Jessica J. Fridrich,et al.  Rich Models for Steganalysis of Digital Images , 2012, IEEE Transactions on Information Forensics and Security.

[18]  Fatih Kurugollu,et al.  A New Methodology in Steganalysis: Breaking Highly Undetectable Steganograpy (HUGO) , 2011, Information Hiding.

[19]  Rainer Bhme.  Advanced Statistical Steganalysis , 2010 .

[20]  Andrew D. Ker.  Steganalysis of Embedding in Two Least-Significant Bits , 2007, IEEE Transactions on Information Forensics and Security.

[21]  Jessica J. Fridrich,et al.  Reliable detection of LSB steganography in color and grayscale images , 2001, MM&Sec '01.

[22]  Tomás Pevný,et al.  Steganalysis by Subtractive Pixel Adjacency Matrix , 2009, IEEE Transactions on Information Forensics and Security.

[23]  Lionel Fillatre.  Adaptive Steganalysis of Least Significant Bit Replacement in Grayscale Natural Images , 2012, IEEE Transactions on Signal Processing.

[24]  Rainer Böhme,et al.  Advanced Statistical Steganalysis , 2010, Information Security and Cryptography.

[25]  Andrew D. Ker.  Fourth-order structural steganalysis and analysis of cover assumptions , 2006, Electronic Imaging.

[26]  Jessica J. Fridrich,et al.  Calibration revisited , 2009, MM&Sec '09.

[27]  Andrew D. Ker.  A Fusion of Maximum Likelihood and Structural Steganalysis , 2007, Information Hiding.

[28]  Sorina Dumitrescu,et al.  LSB steganalysis based on high-order statistics , 2005, MM&Sec '05.

[29]  Tomás Pevný,et al.  "Break Our Steganographic System": The Ins and Outs of Organizing BOSS , 2011, Information Hiding.

[30]  Florent Retraint,et al.  Statistical Decision Methods in Hidden Information Detection , 2011, Information Hiding.