Audio steganalysis using deep belief networks
- 325 Downloads
This paper presents a new steganalysis method that uses a deep belief network (DBN) as a classifier for audio files. It has been tested on three steganographic techniques: StegHide, Hide4PGP and FreqSteg. The results were compared to two other existing robust steganalysis methods based on support vector machines (SVMs) and Gaussian mixture models (GMMs). Afterwards, another classification task aiming at identifying the type of steganographic applied or not to the speech signal was carried out. The results of this four-way classification show that in most cases, the proposed DBN-based steganalysis method gives higher classification rates than the two other steganalysis methods based on SVMs and GMMs.
KeywordsAudio steganography Audio steganalysis DBN MFCCs SVMs GMMs
- Altun, O., Sharma, G., Celik, M. U., Sterling, M., Titlebaum, E. L., & Bocko, M. (2005). Morphological steganalysis of audio signals and the principle of diminishing marginal distortions. In ICASSP, 2, 21–24.Google Scholar
- Garofolo, J. S., et al. (1993). TIMIT: acoustic-phonetic continuous speech corpus LDC93S1. Web download. Philadelphia: Linguistic Data Consortium.Google Scholar
- Ghasemzadeh, H., & Arjmandi, M. K. (2014). Reversed-mel cepstrum based audio steganalysis. In 2014 4th international eConference on computer and knowledge engineering (ICCKE), (pp. 679–684). IEEE.Google Scholar
- Hetzl, S. (2003). StegHide steganography. http://www.steghide.sourceforge.net/.
- Johnson, M. K., Lyu, S., & Farid, H. (2005). Steganalysis of recorded speech. In Proceedings of the electronic imaging 2005, (pp. 664–672). International Society for Optics and Photonics.Google Scholar
- Kraetzer, C., & Dittmann, J. (2007). Mel-cepstrum-based steganalysis for voip steganography. In Proceedings of the electronic imaging 2007. (pp. 664–672) . International Society for Optics and Photonics.Google Scholar
- Ozer, H., Avcibas, I., Sankur, B., & Memon, N. D. (2003). Steganalysis of audio based on audio quality metrics. In Proceedings of the electronic imaging 2003. (pp. 55–66). International Society for Optics and Photonics.Google Scholar
- Palm, R. B. (2012). Prediction as a candidate for learning deep hierarchical models of data. Master’s thesis, Technical University of Denmark.Google Scholar
- Rekik, S., Selouani, S.-A., Guerchi, D., & Hamam, H. (2012). An autoregressive time delay neural network for speech steganalysis. In 2012 11th international conference on information science, signal processing and their applications (ISSPA). (pp. 54–58). IEEE.Google Scholar
- Repp, H. (1996). Hide4PGP Steganography. http://www.heinz-repp.onlinehome.de/Hide4PGP.htm.
- Swanson, E., Ganier, C., Holman, R., & Rosser, J. (2002). Freqency domain steganography. https://www.clear.rice.edu/elec301/Projects01/smokey_steg/group.html.
- Yürüklü, E., Koçal, O. H., & Dilaveroğlu, E. (2014). A new approach for speech audio steganalysis using delay vector variance method. Uludağ Üniversitesi Mühendislik Fakültesi Dergisi, 19(1), 27–36.Google Scholar