Steganalysis of LSB Using Energy Function
This paper introduces an approach to estimate energy of pixel associated with its neighbors. We define an energy function of a pixel which replaces the pixel value by mean or median value of its neighborhood. The correlations inherent in a cover signal can be used for steganalysis, i.e, detection of presence of hidden data. Because of the interpixel dependencies exhibited by natural images this function was able to differentiate between cover and stego image. Energy function was modeled using Gibbs distribution even though pixels in an image have the property of Markov Random Field. Our method is trained to specific embedding techniques and has been tested on different textured images and is shown to provide satisfactory result in classifying cover and stego using energy distribution.
KeywordsMarkov random field Steganography Steganalysis Gibbs distribution
Unable to display preview. Download preview PDF.
- 1.Bouman, C.: Model based image processing. Purdue University (2013)Google Scholar
- 2.Chellappa, R., Jain, A.: Markov random fields. theory and application, vol. 1 (1993)Google Scholar
- 3.Fridrich, J., Goljan, M.: Practical steganalysis of digital images-state of the art. In: Proceedings of SPIE, vol. 4675, pp. 1–13Google Scholar
- 4.Goljan, M., Fridrich, J., Cogranne, R.: Rich model for steganalysis of color images. In: IEEE Workshop on Information Forensic and Security, Atlanta, GA (2014)Google Scholar
- 5.Johnson, N.F., Jajodia, S.: Exploring steganography: seeing the unseen, vol. 31, pp. 26–34. IEEE (1998)Google Scholar
- 7.Krizhevsky, A., Hinton, G.E., et al.: Factored 3-way restricted boltzmann machines for modeling natural images. In: International Conference on Artificial Intelligence and Statistics, pp. 621–628 (2010)Google Scholar
- 8.Li, S.Z.: Markov random field modeling in computer vision. Springer-Verlag New York, Inc. (1995)Google Scholar
- 9.Osindero, S., Hinton, G.E.: Modeling image patches with a directed hierarchy of markov random fields. In: Advances in Neural Information Processing Systems, pp. 1121–1128 (2008)Google Scholar
- 10.Rangarajan, A., Chellappa, R.: Markov random field models in image processing. Citeseer (1995)Google Scholar
- 11.Ranzato, M., Hinton, G.E.: Modeling pixel means and covariances using factorized third-order boltzmann machines. In: 2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2551–2558. IEEE (2010)Google Scholar
- 13.Roth, S., Black, M.J.: Fields of experts: a framework for learning image priors. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, vol. 2, pp. 860–867. IEEE (2005)Google Scholar