Abstract
Steganography is the art and science of producing covert communications by concealing secret messages in apparently innocent media, while steganalysis is the art and science of detecting the existence of these. This manuscript proposes a novel blind statistical steganalysis technique to detect Least Significant Bit (LSB) flipping image steganography. It shows that the technique has a number of major advantages. First, a novel method of pixel color correlativity analysis in Pixel Similarity Weight (PSW). Second, filtering out image pixels according to their statistically detected suspiciousness, thereby excluding neutral pixels from the steganalysis process. Third, ranking suspicious pixels according to their statistically detected suspiciousness and determining the influence of such pixels based on the level of detected anomalies. Fourth, the capability to classify and analyze pixels in three pixel classes of flat, smooth and edgy, thereby enhancing the sensitivity of the steganalysis. Fifth, achieving an extremely high efficiency level of 98.049% in detecting 0.25bpp stego images with only a single dimension analysis.
Similar content being viewed by others
References
Abdullah SM, Manaf AA, Zamani M (2010) Capacity and quality improvement in reversible image watermarking approach. Proceeding - 6th International Conference on Networked Computing and Advanced Information Management, NCM
Altaay AAJ, Sahib SB, Zamani M, (2012) An introduction to image steganography techniques. Proceedings - 2012 International Conference on Advanced Computer Science Applications and Technologies, ACSAT
Amin MM, Salleh M, Ibrahim S, Kitmin MR, Shamsuddin MZI (2003) Information hiding using steganography. in: 4th Natl. Conf. on Telecommunication Technology Proceedings, Shah Alam, Malaysia
Avcibas I, Memon N, Sankur B (2002) Image steganalysis with binary similarity measures. in: IEEE Int. Conf. on Image Processing, Rochester, New York, September 2002
Benton R, Chu H (2005) Soft computing approach to steganalysis of LSB embedding in digital images. in: 3rd Int. Conf. on Information Technology Research and Education, 105–109, 27–30 June 2005
Celik MU, Sharma G, Tekalp AM (2004) Universal image steganalysis using rate-distortion curves. in: Proc. IST/SPIE 16th Annu. Symp. Electronic Imaging Science Technology, San Jose, CA, 19–22, January 2004
Chaeikar SS, Razak SA, Zeidanloo HR, Zamani M, Jaryani F (2010) Interpretative key management (IKM), A novel framework. 2nd International Conference on Computer Research and Development, ICCRD
Chandramouli R, Kharrazi M, Memon N (2003) Image steganography and steganalysis: Concepts and practices. in: Proc. 2nd Int. Workshop Digital Watermarking, Seoul, Korea, 35–49
Chen X-D (2006) Detect LSB steganography with bit plane randomness tests, in: Proc. of 6th World Congress on Intelligent Control and Automation, China, June 21–23, 2006
Cho S, Cha B-H, Wang J, Kuo C-CJ (2010) Block-based image steganalysis: algorithm and performance evaluation. in: Proc. IEEE Int. Symp. Circuits and Systems Paris, France
Cristianini N, Shawe-Taylor J (2000) An introduction in support vector machines and other kernel-based learning methods. Cambridge University Press, New York
Dumitrescu S, Wu X (2005) A new framework of LSB steganalysis of digital media. IEEE Trans Signal Process 53:3936–3947
Dumitrescu S, Wu X, Wang Z (2003) Detection of LSB steganography via sample pair analysis. IEEE Trans. Signal Process, 1995–2007
Fakhredanesh M, Safabakhsh R, Rahmati M (2014) A model-based image steganography method using Watson’s visual model. ETRI J 36:479–489
Fridrich J (1998) Applications of data hiding in digital images, in Melbourne, Tutorial for the ISPACS’98 Conference, Australia, November 1998
Fridrich J (2004) Feature-based steganalysis for JPEG images and its implications for future design of steganographic schemes. in: Proc. Int. Workshop on Information Hiding Toronto, Canada
Fridrich J, Goljan M (2002) Practical steganalysis of digital images-state of the art, in: Proc. SPIE Photonics West, Electronic Imaging, Security and Watermarking of Multimedia Contents, San Jose, CA, vol. 4675, 1–13
Fridrich J, Du R, Meng L (2000) Steganalysis of LSB encoding in color images. in: Proc. IEEE Int. Conf. on Multimedia and Expo, New York, July 31–August 2, 2000
Fridrich J, Goljan M, Du R (2001) Detecting LSB steganography in color and gray-scale images. IEEE Multimedia Magaz, Special Issue on Security, 22–28, October–November 2001
Fridrich J, Goljan M, Soukal D (2003) Higher-order statistical steganalysis of palette images. in: Proc. SPIE, Electronic Imaging, Security, Steganography, and Watermarking of Multimedia Contents V, Santa Clara, CA, vol. 5020, 178–190
Garna J, Brazdil P (1999) Linear tree. Intell Data Anal 3:1–22
Ghebleh M, Kanso A (2014) A robust chaotic algorithm for digital image steganography. Commun Nonlinear Sci Numer Simulat, Elsevier B.V. 19:1898–1907
Guan, Q., Dong, J., Tan, T (2011) An effective image steganalysis method based on neighborhood information of pixels. 18th IEEE International Conference on Image Processing
Holub V, Fridrich J (2013) Random projections of residuals for digital image steganalysis. IEEE Transactions on information forensics and security 8:1996–2006
Johnson N, Jajodia S (1998a) Exploring steganography: seeing the unseen. IEEE Comput 31(2):26–34
Johnson NF, Jajodia S (1998b) Steganalysis of images created using current steganography software. In: Lecture notes in computer science, vol. 1525. Springer-Verlag, Berlin, pp. 273–289
Ker A (2004a) Quantitative evaluation of pairs and RS steganalysis. in: Proc. SPIE, Security, Steganography, Watermarking of Multimedia Contents, San Jose, CA, vol. 5306, 83–97
Ker A (2004b) Improved detection of LSB steganography in grayscale images, in: Proc. Inf. Hiding Workshop, Lecture Notes in Computer Science, vol. 3200, Springer, 97–115
Kerns GJ (2011) Introduction to probability and statistics using R. Youngstown State University, Youngstown
Khosravi MJ, Naghsh-Nilchi AR, (2014) A novel joint secret image sharing and robust steganography method using wavelet. Multimedia Systems 20:215–226 Springer
Kodovský J, Fridrich J (2013) Quantitative Steganalysis Using Rich Models. Media Watermarking, Security, and Forensics, 8665, SPIE-IS&T Electronic Imaging
Lerch-Hostalot D, Megias D (2013) LSB matching steganalysis based on patterns of pixel differences and random embedding. Computers & security 32:192–206
Lerch-Hostalot D, Megías D (2016) Unsupervised steganalysis based on artificial training sets. Eng Appl Artif Intell 50:45–59
Lin E, Woertz E, Kam M (2003) LSB steganalysis using support vector regression, in: Proc. IEEE, SME Information Assurance Workshop
Lu P, Luo X, Tang Q, Shen L (2004) An improved sample pairs method for detection of LSB embedding. in: Proc. 6th Int. Workshop Inf. Hiding, Toronto, ON, Canada, May 2004
Maity SP, Maity S, Sil J, Delpha C (2013) Perceptually adaptive MC-SS image watermarking using GA-NN hybridization in fading gain. Engineering Applications of Artificial Intelligence, 31, 3–14. Elsevier Ltd.
McBride BT, Peterson GL, Gustafson SC (2005) A new blind method for detecting novel steganography. Digit Investig 2(1):50–70
Moerland T (2001) Steganography and Steganalysis. Leiden Institute of Advanced Computing Science
Mohammadi FG, Abadeh MS (2014) Image steganalysis using a bee colony based feature selection algorithm. Engineering Applications of Artificial Intelligence 31, 35–43 Elsevier Ltd.
Nicholas J (1999) Introduction to descriptive statistics. University of Sydney, Sydney
Park TH, Han JG, Moon YH, Eom IK (2016) Performance improvement of LSB-based steganalysis using bit-plane decomposition of images. The Imaging Science Journal 64(5):262–266
Pathak P, Selvakumar S (2014) Blind image steganalysis of JPEG images using feature extraction through the process of dilation. Digit Investig 11(1):67–77
Pevny T, Fridrich J (2007) Merging Markov and DCT features for multi-class JPEG steganalysis. in: Proc. SPIE Conf. Security, Watermarking, and Steganography, San Jose, CA
Provos N, Honeyman P (2001) Detecting steganographic content on the internet. University of Michigan, Techn. Rep. CITI 01-1a
Qiao M, Sung AH, Liu Q (2013) MP3 audio steganalysis. Information Sciences, 231, 123–134, Elsevier Inc
Quinlan JR (1993) C4.5: programs for machine learning. Morgan Kaufmann, San Mateo
Rodriguez B, Peterson G, Bauer K, Agaian S (2006) Steganalysis embedding percentage determination with learning vector quantization. in: Proc. IEEE Int. Conf. Systems Man and Cybernetics, Taipei, Taiwan
Roue B, Bas P, Chassery J-M (2004) Improving LSB steganalysis using marginal and joint probabilistic distributions. in: Proc. 6th Workshop on Multimedia & Security, MM&Sec 2004, Magdeburg, Germany, September 20–21, 2004, ACM
Shi Y, Chen C, Chen W (2006) A Markov process based approach to effective attacking JPEG steganography. in: Proc. Int. Workshop on Information Hiding, Old Town Alexandria, VA
Smola AJ, Scholkopf B (1998) A tutorial on support vector regression. Techn. Rep. NC2-TR-1998 030, October 1998
Stallings W (2004) Cryptography and network security: principles and practices, fourth edn. Dorling Kindersley (Pearson Education, Pvt. Ltd.), India
Thai TH, Cogranne R, Retraint F (2014) Statistical Model of Quantized DCT Coefficients: Application in the Steganalysis of Jsteg Algorithm. IEEE Transactions on image processing, IEEE 1980–1993.
Tingting Z, Lina W, Yu F, Yanzhen R (2013) JPEG Steganalysis Based on Locality Preserving Projection Dimensionality Reduction Method. Applied Mechanics and Materials 411–414:1185–1188 Trans Tech Publications
Westfeld A (2002) Detecting low embedding rates. in: Lecture Notes in Computer Science, vol. 2578, Springer-Verlag, Berlin, 324–339
Westfeld A, Pfitzmann A (1999) Attacks on steganographic systems. In: Proc. of Information Hiding, Third Int. Workshop, Dresden, Germany, September 28–October 1, 61–75
Zamani M, Manaf ABA (2015) Genetic algorithm for fragile audio watermarking. Telecommunication Systems
Zamani M, Taherdoost H, Manaf AA, Ahmad RB, Zeki AM (2009a) Robust audio steganography via genetic algorithm. 2009 International Conference on Information and Communication Technologies, ICICT
Zamani M, Manaf AA, Ahmad R (2009b) Current problems of substitution techniques of audio steganography. International Conference on Artificial Intelligence and Pattern Recognition 2009, AIPR
Zamani M, Manaf ABA, Ahmad RB, Jaryani F, Chaeikar SS, Zeidanloo HR (2010) Genetic audio watermarking. Communications in Computer and Information Science
Zamani M, Manaf ABA, Zeidanloo HR, Chaeikar SS (2011) Genetic substitution-based audio steganography for high capacity applications. International Journal of Internet Technology and Secured Transactions
Zamani M, Manaf ABA, Abdullah SM (2012a) An overview on audio steganography techniques. International Journal of Digital Content Technology and its Applications
Zamani M, Abdul Manaf AB Daruis R (2012b) Azizah technique for efficiency measurement in steganography. Proceedings - ICIDT 2012, 8th International Conference on Information Science and Digital Content Technology
Zamani M, Azizah BAM, Shahidan MA, Shojae CS (2012c) Mazdak technique for PSNR estimation in audio steganography. Applied Mechanics and Materials
Zhang T, Ping X (2003) Reliable detection of LSB steganography based on difference image histogram. in: Proc. ICASSP, vol. I, 545–548
Zhang H, Ping X, (2013) Multi-class Markov models for JPEG steganalysis. Proceedings of Fifth International Conference on Digital Image Processing (ICDIP 2013). Beijing, China
Zhi L, Fen SA, Xian YY (2003) A LSB steganography detection algorithm. in: Proc. IEEE Int. Symp. on Personal, Indoor and Mobile Radio Communication
Zong H, Liu F-l, Luo X-y (2012) Blind image steganalysis based on wavelet coefficient correlation. Digit Investig 9(1):58–68
Luo W, Huang F, Huang J (2010) Edge adaptive image steganography based on LSB matching revisited. IEEE transaction on information forensics and security 5(2):201–214
Acknowledgements
This work is a part of wider research supported by the Universiti Teknologi Malaysia.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix A – Pixel classification sample
After an initial PSW analysis of the image pixels, neutral pixels are excluded from the subsequent steganalysis process and suspicious pixels are categorized into three pixel classes of flat, smooth and edgy. Figure 12 below illustrates the performance of the PSW steganalysis in doing this.
Appendix B – Pixel reference profiles
Pixel reference profiles are a summary of the purified, classified and prioritized PSW analysis results of the SVM training database. The pixel reference profile with the highest matching score with the analyzed image helps to determine the results of the steganalysis in detecting embedded messages and estimating message length.
PSW steganalysis uses three pixel reference profiles: cover, 0.125bpp and 0.25bpp. The cover pixel reference profile was constructed from the PSW analysis results on cover images in the SVM training database, while the 0.125bpp and 0.25bpp pixel reference profiles were created from the PSW analysis results on 0.125bpp and 0.25bpp stego images respectively in the SVM training database. Tables 10, 11 and 12 present the cover, 0.125bpp and 0.25bpp pixel reference profiles respectively. In these tables, SD stands for standard deviation, S stands for achieved matching score, and x is the PSW value.
Rights and permissions
About this article
Cite this article
Shojae Chaeikar, S., Zamani, M., Abdul Manaf, A.B. et al. PSW statistical LSB image steganalysis. Multimed Tools Appl 77, 805–835 (2018). https://doi.org/10.1007/s11042-016-4273-6
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-016-4273-6