Structural Design of Convolutional Neural Networks for Steganalysis

Recent studies have indicated that the architectures of convolutional neural networks (CNNs) tailored for computer vision may not be best suited to image steganalysis. In this letter, we report a CNN architecture that takes into account knowledge of steganalysis. In the detailed architecture, we take absolute values of elements in the feature maps generated from the first convolutional layer to facilitate and improve statistical modeling in the subsequent layers; to prevent overfitting, we constrain the range of data values with the saturation regions of hyperbolic tangent (TanH) at early stages of the networks and reduce the strength of modeling using 1×1 convolutions in deeper layers. Although it learns from only one type of noise residual, the proposed CNN is competitive in terms of detection performance compared with the SRM with ensemble classifiers on the BOSSbase for detecting S-UNIWARD and HILL. The results have implied that well-designed CNNs have the potential to provide a better detection performance in the future.

[1]  Yi Zhang,et al.  Steganalysis of Adaptive JPEG Steganography Using 2D Gabor Filters , 2015, IH&MMSec.

[2]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Jessica J. Fridrich,et al.  Random Projections of Residuals for Digital Image Steganalysis , 2013, IEEE Transactions on Information Forensics and Security.

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

[5]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[6]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

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

[8]  Bin Li,et al.  Stacked convolutional auto-encoders for steganalysis of digital images , 2014, Signal and Information Processing Association Annual Summit and Conference (APSIPA), 2014 Asia-Pacific.

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

[10]  Pedro Comesaña Alfaro,et al.  Improving Selection-Channel-Aware Steganalysis Features , 2016, Media Watermarking, Security, and Forensics.

[11]  Jiwu Huang,et al.  Adaptive steganalysis against WOW embedding algorithm , 2014, IH&MMSec '14.

[12]  Yun Q. Shi,et al.  Textural Features for Steganalysis , 2012, Information Hiding.

[13]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[14]  Jessica J. Fridrich,et al.  Steganalysis of JPEG images using rich models , 2012, Other Conferences.

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

[16]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  Michael S. Bernstein,et al.  ImageNet Large Scale Visual Recognition Challenge , 2014, International Journal of Computer Vision.

[18]  Bin Li,et al.  A new cost function for spatial image steganography , 2014, 2014 IEEE International Conference on Image Processing (ICIP).

[19]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[20]  Jessica J. Fridrich,et al.  Designing steganographic distortion using directional filters , 2012, 2012 IEEE International Workshop on Information Forensics and Security (WIFS).

[21]  Trevor Darrell,et al.  Caffe: Convolutional Architecture for Fast Feature Embedding , 2014, ACM Multimedia.

[22]  Jessica J. Fridrich,et al.  Selection-channel-aware rich model for Steganalysis of digital images , 2014, 2014 IEEE International Workshop on Information Forensics and Security (WIFS).

[23]  Jessica J. Fridrich,et al.  Low-Complexity Features for JPEG Steganalysis Using Undecimated DCT , 2015, IEEE Transactions on Information Forensics and Security.

[24]  Bin Li,et al.  Investigation on Cost Assignment in Spatial Image Steganography , 2014, IEEE Transactions on Information Forensics and Security.

[25]  Yun Q. Shi,et al.  Variable Multi-dimensional Co-occurrence for Steganalysis , 2014, IWDW.

[26]  Jessica J. Fridrich,et al.  Content-Adaptive Steganography by Minimizing Statistical Detectability , 2016, IEEE Transactions on Information Forensics and Security.

[27]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[28]  Tomás Pevný,et al.  Using High-Dimensional Image Models to Perform Highly Undetectable Steganography , 2010, Information Hiding.

[29]  Jiwu Huang,et al.  Adaptive Steganalysis Based on Embedding Probabilities of Pixels , 2016, IEEE Transactions on Information Forensics and Security.

[30]  Jessica J. Fridrich,et al.  Universal distortion function for steganography in an arbitrary domain , 2014, EURASIP Journal on Information Security.

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

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

[33]  Z. Jane Wang,et al.  Median Filtering Forensics Based on Convolutional Neural Networks , 2015, IEEE Signal Processing Letters.

[34]  Jing Dong,et al.  Deep learning for steganalysis via convolutional neural networks , 2015, Electronic Imaging.

[35]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[36]  Marc Chaumont,et al.  Deep learning is a good steganalysis tool when embedding key is reused for different images, even if there is a cover sourcemismatch , 2015, Media Watermarking, Security, and Forensics.