The objective of Image steganalysis is the detection of presence of hidden content in any given image. Steganalysis is a binary classification problem for classifying a given image into one of two classes either Stego or Cover. Conventional Steganalysis consisted of a two step method, feature extraction followed by classification using machine learning. This feature extraction process required an in-depth knowledge of image statistics which are affected by hiding the secret data. With the advent of Deep Learning, Convolution neural networks(CNN) are being widely used for image classification, with an advantage of automatic feature learning. CNN based Steganalysis methods have made the feature extraction step simple as the steganalyzer does not need to specify the features which are affected by data hiding. Added to this feature extraction step and classification step are integrated into a single step. In this paper we have reviewed the existing CNN based steganalysis methods and proposed a novel CNN architecture customized for the task of steganalysis named StegNet. StegNet is built based on deep residual learning. And each feature map is assigned a weight to determine the priority by using global average pooling.

Steganalysis, feature leaning, CNN, Steganography, residual learning

[1] I. Cox, M. Miller, J. Bloom, J. Fridrich, and T. Kalker, Digital water- marking and steganography. Morgan kaufmann, 2007.
[2] J. Fridrich, Steganography in digital media: principles, algorithms, and applications. Cambridge University Press, 2009.
[3] J. Fridrich, M. Goljan, and R. Du, “Detecting lsb steganography in color, and gray-scale images,” IEEE multimedia, vol. 8, no. 4, pp. 22–28, 2001.
[4] I. Avcibas, N. Memon, and B. Sankur, “Steganalysis using image qual- ity metrics,” IEEE transactions on Image Processing, vol. 12, no. 2, pp. 221–229, 2003.
[5] S. Lyu and H. Farid, “Detecting hidden messages using higher-order statistics and support vector machines,” in International Workshop on Information Hiding, pp. 340–354, Springer, 2002.
[6] S. Liu, H. Yao, and W. Gao, “Steganalysis based on wavelet texture analysis and neural network,” in Intelligent Control and Automation, 2004. WCICA 2004. Fifth World Congress on, vol. 5, pp. 4066–4069, IEEE, 2004.
[7] H. Farid, “Detecting hidden messages using higher-order statistical mod- els,” in Image Processing. 2002. Proceedings. 2002 International Con- ference on, vol. 2, pp. II–II, IEEE, 2002.
[8] S. Lyu and H. Farid, “Steganalysis using higher-order image statistics,” IEEE transactions on Information Forensics and Security, vol. 1, no. 1, pp. 111–119, 2006.
[9] S. Lyu and H. Farid, “Steganalysis using color wavelet statistics and one- class support vector machines,” in Security, Steganography, and Water- marking of Multimedia Contents VI, vol. 5306, pp. 35–46, International Society for Optics and Photonics, 2004.
[10] W.-N. Lie and G.-S. Lin, “A feature-based classification technique for blind image steganalysis,” IEEE transactions on multimedia, vol. 7, no. 6, pp. 1007–1020, 2005.
[11] Y. Wang and P. Moulin, “Optimized feature extraction for learning- based image steganalysis,” IEEE Transactions on Information Forensics and Security, vol. 2, no. 1, pp. 31–45, 2007.
[12] X. Luo, F. Liu, S. Lian, C. Yang, and S. Gritzalis, “On the typical statistic features for image blind steganalysis,” IEEE Journal on selected areas in Communications, vol. 29, no. 7, pp. 1404–1422, 2011.
[13] S.-H. Zhan and H.-B. Zhang, “Blind steganalysis using wavelet statistics and anova,” in Machine Learning and Cybernetics, 2007 International Conference on, vol. 5, pp. 2515–2519, IEEE, 2007.
[14] X. Luo, F. Liu, J. Chen, and Y. Zhang, “Image universal steganalysis based on wavelet packet transform,” in Multimedia Signal Processing, 2008 IEEE 10th Workshop on, pp. 780–784, IEEE, 2008.
[15] J. Kodovsky`, J. J. Fridrich, and V. Holub, “Ensemble classifiers for steganalysis of digital media.,” IEEE Trans. Information Forensics and Security, vol. 7, no. 2, pp. 432–444, 2012.
[16] F. Li, X. Zhang, B. Chen, and G. Feng, “Jpeg steganalysis with high- dimensional features and bayesian ensemble classifier,” IEEE signal pro- cessing letters, vol. 20, no. 3, pp. 233–236, 2013.
[17] Z. Sun, M. Hui, and C. Guan, “Steganalysis based on co-occurrence ma- trix of differential image,” in Intelligent Information Hiding and Mul- timedia Signal Processing, 2008. IIHMSP’08 International Conference on, pp. 1097–1100, IEEE, 2008.
[18] Z. Sun, M. Hui, and C. Guan, “Steganalysis based on co-occurrence ma- trix of differential image,” in Intelligent Information Hiding and Mul- timedia Signal Processing, 2008. IIHMSP’08 International Conference on, pp. 1097–1100, IEEE, 2008
[19] Q.-l. Deng and J.-j. Lin, “A universal steganalysis using features derived from the differential image histogram in frequency domain,” in Image and Signal Processing, 2009. CISP’09. 2nd International Congress on, pp. 1–4, IEEE, 2009.
[20] J. Dong and T. Tan, “Blind image steganalysis based on run-length histogram analysis,” in Image Processing, 2008. ICIP 2008. 15th IEEE International Conference on, pp. 2064–2067, IEEE, 2008.
[21] I. Avciba¸s, M. Kharrazi, N. Memon, and B. Sankur, “Image steganalysis with binary similarity measures,” EURASIP Journal on Applied Signal Processing, vol. 2005, pp. 2749–2757, 2005.
[22] H. Sajedi and M. Jamzad, “A steganalysis method based on contourlet transform coefficients,” in Intelligent Information Hiding and Multime- dia Signal Processing, 2008. IIHMSP’08 International Conference on, pp. 245–248, IEEE, 2008
[23] M. Sheikhan, M. S. Moin, and M. Pezhmanpour, “Blind image steganal- ysis via joint co-occurrence matrix and statistical moments of contourlet transform,” in Intelligent Systems Design and Applications (ISDA), 2010 10th International Conference on, pp. 368–372, IEEE, 2010.
[24] T. Pevny and J. Fridrich, “Merging markov and dct features for multi- class jpeg steganalysis,” in Security, Steganography, and Watermarking of Multimedia Contents IX, vol. 6505, p. 650503, International Society for Optics and Photonics, 2007.
[25] T. Pevny, P. Bas, and J. Fridrich, “Steganalysis by subtractive pixel adjacency matrix,” IEEE Transactions on information Forensics and Security, vol. 5, no. 2, pp. 215–224, 2010.
[26] J. Fridrich and J. Kodovsky, “Rich models for steganalysis of digital im- ages,” IEEE Transactions on Information Forensics and Security, vol. 7, no. 3, pp. 868–882, 2012.
[27] V. Holub and J. Fridrich, “Random projections of residuals for digital image steganalysis,” IEEE Transactions on Information Forensics and Security, vol. 8, no. 12, pp. 1996–2006, 2013.
[28] Y. Qian, J. Dong, W. Wang, and T. Tan, “Deep learning for steganalysis via convolutional neural networks,” in Media Watermarking, Security, and Forensics 2015, vol. 9409, p. 94090J, International Society for Optics and Photonics, 2015.
[29] Y. Qian, J. Dong, W. Wang, and T. Tan, “Feature learning for ste- ganalysis using convolutional neural networks,” Multimedia Tools and Applications, pp. 1–25, 2017.
[30] W. Rawat and Z. Wang, “Deep convolutional neural networks for image classification: A comprehensive review,” Neural computation, vol. 29, no. 9, pp. 2352–2449, 2017.
[31] S. Tan and B. Li, “Stacked convolutional auto-encoders for steganalysis of digital images,” in Signal and Information Processing Association Annual Summit and Conference (APSIPA), 2014 Asia-Pacific, pp. 1–4, IEEE, 2014.
[32] T. Pevny`, T. Filler, and P. Bas, “Using high-dimensional image models to perform highly undetectable steganography,” in International Work- shop on Information Hiding, pp. 161–177, Springer, 2010.
[33] V. Holub and J. Fridrich, “Designing steganographic distortion using directional filters,” in 2012 IEEE International workshop on information forensics and security (WIFS), pp. 234–239, IEEE, 2012.
[34] V. Holub and J. Fridrich, “Digital image steganography using universal distortion,” in Proceedings of the first ACM workshop on Information hiding and multimedia security, pp. 59–68, ACM, 2013.
[35] G. Xu, H.-Z. Wu, and Y.-Q. Shi, “Structural design of convolutional neu- ral networks for steganalysis,” IEEE Signal Processing Letters, vol. 23, no. 5, pp. 708–712, 2016.
[36] Y. Qian, J. Dong, W. Wang, and T. Tan, “Learning and transferring rep- resentations for image steganalysis using convolutional neural network,” in Image Processing (ICIP), 2016 IEEE International Conference on, pp. 2752–2756, IEEE, 2016.
[37] J.-F. Couchot, R. Couturier, C. Guyeux, and M. Salomon, “Steganalysis via a convolutional neural network using large convolution filters for em- bedding process with same stego key,” arXiv preprint arXiv:1605.07946,
[38] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 770–778, 2016.
[39] J. Hu, L. Shen, and G. Sun, “Squeeze-and-excitation networks,” in Pro- ceedings of the IEEE conference on computer vision and pattern recog- nition, pp. 7132–7141, 2018.
[40] J. B. Guttikonda and R. Sridevi, “A new steganalysis approach with an efficient feature selection and classification algorithms for identifying the stego images,” Multimedia Tools and Applications, pp. 1–19, 2019.
[41] S. Wu, S. Zhong, and Y. Liu. “Deep residual learning for image steganalysis”, Multimedia tools and applications, 77(9):10437–10453, 2018.