This paper proposes a novel Universal Steganalysis framework which can be applied for spatial domain and JPEG Domain Steganography algorithms. The objective is to develop a steganalysis algorithm which has to identify any distribution (uniform or non-uniform) of stego-payloads. The framework proposed uses a 3-way tensor model to extract the image features which is important for estimating the embedded change irrespective of domain. To obtain the accurate results and to analyze the error, 360 degree bit change estimation is done. The experimental results evaluated on 3000 images which shows a good detection rate in both domains and a reasonable false acceptance rate and false rejection rate based on the pay load when tested with most of the steganography algorithms.

Spatial Domain, Jpeg Domain, Steganalysis, Tensor, SVM

[1] Jan Kodovsk� and Jessica Fridrich, �Quantitative Steganalysis Using Rich Models�, Proc. SPIE 8665, Media Watermarking, Security, and Forensics, March 22, 2013.
[2] J. Fridrich and J. Kodovsk�. Rich models for steganalysis of digital images. IEEE Transactions on Information Forensics and Security, 7(3):868�882, June 2012.
[3] J. Kodovsk�, J. Fridrich, and V. Holub, �Ensemble classifiers for steganalysis of digital media�, IEEE Transactions on Information Forensics and Security , 7(2):432�444, April 2012
[4] T. Pevn�, J. Fridrich, and A. Ker, �From blind to quantitative steganalysis,� IEEE Trans. Inf. Forensics Security, vol. 7, no. 2, pp. 445�454, Apr. 2012.
[5] Chunfang Yang, Fenlin Liu, Xiangyang Luo, and Ying ZengPixel, �Pixel Group Trace Model-Based Quantitative Steganalysis for Multiple Least-Significant Bits Steganography�, IEEE Transactions On Information Forensics And Security, vol. 8, no. 1, January 2013
[6] Chunfang Yang, Fenlin Liu1, Xiangyang Luo1, Ying Zeng,�Fusion of Two Typical Quantitative Steganalysis Based on SVR�, Journal Of Software, Vol. 8, No. 3, March 2013
[7] Tomas Pevny and Andrew D. Ker, �The Challenges of Rich Features in Universal Steganalysis�, Proc. SPIE 8665, Media Watermarking, Security, and Forensics, 86650M, March 22, 2013.
[8] Ziwen Sun, Hui Li, �Quantitative Steganalysis Based on Wavelet Domain HMT and PLSR�, 10th IEEE International Symposium on Distributed Computing and Applications to Business, Engineering and Science, 2011
[9] Gikhan Gul and Faith Kurugollu,�SVD-Based Universal Spatial Domain Image Steganalysis�, IEEE Transactions on Information Forensics and Security,Vol.5,No.2,June 2010
[10] Changxin Liu, Chunjuan Oujang,�Image Steganalysis Based on Spatial Domain and DWT Domain Features�, IEEE Second International Conference on Network Security, Wireless Communications and Trusted Computing, 2010.
[11] Souvik Bhattacharyya and Gautam Sanyal, �Steganalysis of LSB Image Steganography using Multiple Regression and Auto Regressive (AR) Model�, Int. J. Comp.Tech. Appl., Vol 2 (4), 1069-1077
[12] Zhenhao Zhu, Tao Zhang, Baoji Wan, �A special detector for the edge adaptive image steganography based on LSB matching revisited�, 2013 10th IEEE International Conference on Control and Automation (ICCA)
[13] Weiqi Luo, Yuangen Wang, Jiwu Huang, �Security analysis on spatial +_1Steganography for jpeg decompressed images�, IEEE Signal Processing Letters, (Volume:18 ,Issue: 1)
[14] Tomas Pevny,Tomas Filler, Patrick Bas, �Using high-dimentional image models to perform undetectable steganography� , Lecture Notes in Computer Science Volume 6387,2010,pp 161-177
[15] Mielikainen, � LSB Matching revisited�, Signal Processing Letters, IEEE (Volume:13 , Issue: 5), May 2006
[16] Weiqi Luo, Fangjun Huang, and Jiwu Huang, �Edge Adaptive Image Steganography Based on LSB Matching Revisited�, IEEE Transactions On Information Forensics And Security, vol. 5, no. 2, june 2010
[17] Shunquan Tan, �Targeted Steganalysis of Edge Adaptive Image Steganography Based on LSB Matching Revisited Using B-Spline Fitting�, Signal Processing Letters, IEEE(Volume:19 ,Issue: 6), June 2012
[18] Chao Wang, Xiaolong Li, Bin Yang, Xiaoqing Lu, Chengcheng Liu, � Content- adaptive approach for reducing embedding impact insteganography�, IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 14-19 March 2010
[19] Yi-zhen Chen, Zhi Han, Shu-ping Li, Chun-hui Lu, Xiao-Hui Yao, �An Adaptive Steganography algorithm based on block sensitivity vectors using HVS features� 3rd International Congress on Image and Signal Processing (CISP) 16-18 Oct. 2010
[20] Brett w. Bader and Tamara g. Kolda, �MATLAB Tensor Classes for Fast Algorithm Prototyping� ACM Transactions on Mathematical Software, Volume 32 Issue 4, December 2006 Pages 635-653
[21] De Lathauwer, L., De Moor, B., and Vandewalle, J 2000a, �A multilinear singular value decomposition� SIAM J. Matrix Anal. Appl. 21, 4, 1253�1278, 2000.
[22] De Lathauwer, L., De Moor, B., and Vandewalle, J. 2000b, � On the best rank-1 and rank- (R1, R2, . . . , RN ) approximation of higher-order tensors�. SIAM J. Matrix Anal. Appl. 21, 4, 1324�1342, 2000.
[23] Kiers, H. A. L, �Towards a standardized notation and terminology in multiway analysis� .J. Chemometrics 14, 105�122., 2000
[24] The MathWorks, Inc. 2004a. Documentation: MATLAB: Programming: Classes and objects http://www.mathworks.com/access/helpdesk/help/techdoc/matlabprog/ch11 mat.html .
[25] The MathWorks, Inc. 2004b. Documentation: MATLAB: Programming: Multidimensional arrays. http://www.mathworks.com/access/helpdesk/help/techdoc/matlab prog/ch dat32.html#39663 .
[26] www.sandia.gov/~tgkolda/TensorToolbox/
[27] P. Sallee, �Model-based steganography,� in Digital Watermarking, 2nd International Workshop, ser. Lecture Notes in Computer Science, T. Kalker, I. J. Cox, and Y. M. Ro, Eds., vol. 2939. Seoul, Korea: Springer-Verlag, New York, October 20�22, 2003, pp. 154�167.
[28] D. Upham, http://zooid.org/~paul/crypto/jsteg/.
[29] Guangjie Liu, Zhan Zhang and Yuewei Dai, �Improved LSB-matching Steganography for Preserving Second-order Statistics�, Journal of Multimedia, vol. 5, no. 5, October 2010
[30] J. Fridrich and D. Soukal, �Matrix embedding for large payloads,� in Proceedings SPIE, Electronic Imaging, Security, Steganography, and Watermarking of Multimedia Contents VIII, E. J. Delp and P. W. Wong, Eds., vol. 6072, San Jose, CA, January 16�19, 2006, pp. W1�W10.
[31] A. Latham, http://linux01.gwdg.de/~alatham/stego.html.
[32] Y. Kim, Z. Duric, and D. Richards, �Modified matrix encoding technique for minimal distortion steganography,� in Information Hiding, 8th International Workshop, ser. Lecture Notes in Computer Science, J. L. Camenisch, C. S. Collberg, N. F. Johnson, and P. Sallee, Eds., vol. 4437. Alexandria, VA: Springer-Verlag, New York, July 10�12, 2006, pp. 314�327.
[33] J. Fridrich, T. Pevn�, and J. Kodovsk�, �Statistically undetectable JPEG steganography: Dead ends, challenges, and opportunities,� in Proceedings of the 9th ACM Multimedia & Security Workshop, J. Dittmann and J. Fridrich, Eds., Dallas, TX, September 20�21, 2007, pp. 3�14.