Volume 15, Issue 3 (12-2018)                   JSDP 2018, 15(3): 113-122 | Back to browse issues page


XML Persian Abstract Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

Ashari F, Riahi N. Audio Steganalysis based on Inter-frame correlation and recursive feature elimination. JSDP. 2018; 15 (3) :113-122
URL: http://jsdp.rcisp.ac.ir/article-1-625-en.html
Abstract:   (86 Views)

Dramatic changes in digital communication and exchange of image, audio, video and text files result in a suitable field for interpersonal transfers of hidden information. Therefore, nowadays, preserving channel security and intellectual property and access to hidden information make new fields of researches naming steganography, watermarking and steganalysis. Steganalysis as a binary classification distinguish clean signals from stego signals. Features extracted from time and transform domain are proper for this classifier.
Some of steganalysis methods are depended on a specific steganography algorithm and others are independent. The second group of methods are called Universal steganalysis. Universal steganalysis methods are widely used in applications because of their independency to steganography algorithms. These algorithms are based on characteristics such as distortion measurements, higher order statistics and other similar features.
In this research we try to achieve more reliable and accurate results using analytical review of features, choose more effective of them and optimize SVM performance.
In new researches Mel Frequency Cepstral Coefficient and Markov transition probability matrix coefficients are used to steganalysis design. In this paper we consider two facts. First, MFCC extract signal   features in transform domain similar to human hearing model, which is more sensitive to low frequency signals. As a result, in this method there is more hidden information mostly in higher frequency audio signals. Therefore, it is suggested to use reversed MFCC. Second, there is an interframe correlation in audio signals which is useful as an information hiding effect.
For the first time, in this research, this features is used in steganalysis field. To have more accurate and stable results, we use recursive feature elimination with correlation bias reduction for SVM.
To implement suggested algorithm, we use two different data sets from TIMIT and GRID. For each data sets,Steghide and LSB-Matching steganography methods implement with 20 and 50 percent capacity. In addition, one of the LIBSVM 3.2 toolboxes is sued for implementation.
Finally, the results show accuracy of steganalysis, four to six percent increase in comparison with previous methods. The ROC of methods clearly shows this improvement.
 

Full-Text [PDF 4125 kb]   (49 Downloads)    
Type of Study: Research | Subject: Paper
Received: 2016/12/23 | Accepted: 2018/07/25 | Published: 2018/12/19 | ePublished: 2018/12/19

References
1. [1] Ghasemzadeh, H. Tajikkhas, M. Khalilarjmandi, H. "Audio Steganalysis based on psychoacoustic model of human hearing". (2016) ELSEVIER, Signal Processing Letters,Vol. 51, Pages 664-672.
2. [2] Ren, Y. Xiong, Q. Wang, L. "Steganalysis of AAC using calibrated Markov model of adjacent codebook". (2016) IEEE International Conference on Acoustica, Speech and Signal Processing.
3. [3] Yan, Ke. Zhang, David. "Feature selection and analysis on correlated gas sensor data with-recursive feature elimination". (2015) Elsevier, Sensors and Actuators B 212. Pages 353-363.
4. [4] Ghasemzadeh, H. Khalil Arjmandi, M. "Reversed-Mel Cepstrum Based Audio Steganalysis". (2014) IEEE 4th International eConference on Computer and Knowledge Engineering.
5. [5] Yang, Y. et al. "An inter-frame correlation based error concealment of immittance spectral coeffi-cients for mobile speech and audio codecs". (2014) IEEE International Conference on High Per-formance Computing and Communications.
6. [6] Liu, Q. Sung, A. Qiao, M. "Derivative-Based Audio Steganalysis". (2011) ACM Transactions on Multimedia Computing, Communications and Applications, Vol. 7, No. 3, Article 18.
7. [7] Tolosi, L. Lengauer, Th. "Classification with correlated features: unreliability of feature ranking and solu-tions". (2011) Data and text mining. Publ-ished by Oxford University Press. Vol. 27, Pages 1986–1994.
8. [8] Liu, Q. Sung, A. Qiao, M."Temporal Derivative-Based Spectrum and Mel-Cepstrum Audio Ste-ganalysis". (2009) IEEE Transaction on in-formation forensics and security, Vol. 4, No. 3.
9. [9] Liu, Q. et al. "Novel Stram Mining for Audio Steganalysis". (2009) ACM International Con-ference on Multimedia, Pages 95-104. [PMCID]
10. [10] Zeng, W. et al. "An Algorithm of Echo Steg-analysis based on Power Cepstrum and Pattern Classification". (2008) International Conference on Information and Automation, Pages 1667-1670. [PMID] [PMCID]
11. [11] Liu, Y. el al. "A Novel Audio Steganalysis based on Higher-Order Statistics of Distortion Measure with Hausdroff Distence". (2008) Lecture Notes in Computer Science, Vol. 5222, Pages 487-501.
12. [12] Ismail Avcibas,"Audio Steganalysis With Content-Independent Dis-tortion Measures". (2006) IEEE Signal Processing Letters,Vol.13, No.2.
13. [13] Johnson, M.K. et al. "Steganalysis of Recorded Speech". (2005) Conference on Security, Stega-nography and Watermarking of Multi-media, Vol. 5681, Pages 664-672. [DOI:10.1117/12.586941]
14. [14] Ozer, H. Activbas, I. et al. "Steganalysis of Audio based on Audio Quality Metrics". (2003) Conference on Security, Steganography and Watermarking of Multimedia, Vol. 5020, Pages 55-66. [DOI:10.1117/12.477313]
15. [15] Harmsen, J.J. Pearlman, W.A. "Steganalysis of additive noise modelable information Hiding". (2006)
16. [16] Paliwal, K.K. "Use of temporal correlation between successive frames in a hidden markov-model based speech recognizer". (1993) IEEE international conference on Acoustics, speech, and signal processing
17. [17] I. Guyon, J. Weston, S. B, and V. V. (2002) "Gene selection for cancer classification using support vector machines". Mach. Learn., Vol. 46, No. 1–3, pp. 389–422. [DOI:10.1023/A:1012487302797]

Add your comments about this article : Your username or Email:
CAPTCHA code

Send email to the article author


© 2015 All Rights Reserved | Signal and Data Processing