Efficient stego key recovery based on distribution differences of extracting message bits
- 46 Downloads
The extraction of embedded messages, i.e., extraction attacks are the ultimate purpose of steganalysis, with great practical significance to obtain covert communication content and covert communication forensics. For steganography using a stego key, the extraction attacks are equivalent to the stego key recovery. This paper mainly studies the methods of efficient stego key recovery for LSB steganography in JPEG domain. According to the distribution differences of extracting message bits generated by the correct and incorrect keys, stego key recovery is transformed into the hypothesis test of the message bits distribution extracted by the correct and incorrect keys. Firstly, by fitting the message bits distribution extracted by the correct key, a stego key recovery method based on nonparametric hypothesis test is proposed. Secondly, by utilizing the distribution differences between message bits extracted by the correct and incorrect keys, a stego key recovery method based on parametric hypothesis test is proposed. And then, formulas are given for calculating the sample size and threshold in the proposed stego key recovery model, on the basis of type I error and type II error. Experimental results show that the proposed method can successfully recover the stego key of OutGuess 0.13b, OutGuess0.2, JPEG domain random LSB matching steganography and random F3 steganography, the performance is superior to that of existing methods.
KeywordsSteganalysis Extraction attacks JPEG domain LSB steganography Stego key recovery
This work was supported by the National Natural Science Foundation of China (Grant no. U1636219, U1736214, 61572052, 61602508, and 61772549), the National Key R&D Program of China (Grant no. 2016YFB0801303 and 2016QY01W0105), Plan for Scientific Innovation Talent of Henan Province (no. 2018JR0018), and the Key Technologies R&D Program of Henan Province (Grant no.162102210032).
- 6.Upham, D.: JPEG-JSteg-V4, steganographic algorithm JSteg. http://zooid.org/paul/crypto/jsteg
- 7.Provos, N.: Defending against statistical steganalysis. Proceedings of the 10th conference on USENIX Security Symposium-Volume 10, USENIX Association, pp. 24–24 (2001)Google Scholar
- 8.Westfeld, A.: F5-A steganographic algorithm. Proceedings of the 4th International Workshop on Information Hiding, Springer-Verlag, pp 289–302 (2001)Google Scholar
- 10.Holub, V., Fridrich, J.: Digital image steganography using universal distortion. Proceedings of the 1st ACM Workshop on Information Hiding and Multimedia Security, pp. 59–68 (2013)Google Scholar
- 12.Denemark, T., Fridrich, J.: Model based steganography with precover. Electron. Imaging. 2017(7), 56–66 (2017)Google Scholar
- 14.Fridrich, J., Goljan, M., Soukal, D.: Searching for the stego-key. Proceedings of SPIE 5306, Security, Steganography, and Watermarking of Multimedia Contents VI, San Jose, pp. 70–82 (2004)Google Scholar
- 15.Pevný, T., Ker, A.D.: Steganographic key leakage through payload metadata. Proceedings of the 2nd ACM Workshop on Information Hiding and Multimedia Security, pp. 109–114 (2014)Google Scholar
- 16.Fridrich, J., Goljan, M., Soukal, D., Holotyak, T.: Forensic steganalysis: determining the stego key in spatial domain steganography. Proceedings of SPIE 5681, Security, Steganography, and Watermarking of Multimedia Contents VII, San Jose, pp. 631–642 (2005)Google Scholar
- 18.Chen, J.Y., Zhu, Y.F., Zhang, W.M., Liu, J.F.: Chosen-key extracting attack to random LSB steganography. J. Commun. 31(5), 73–80 (2010)Google Scholar