KIISE Transactions on Computing Practices (정보과학회 컴퓨팅의 실제 논문지)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지
DOI QR코드

DOI QR Code

Generalized Steganalysis using Deep Learning

딥러닝을 이용한 범용적 스테그아날리시스

  • 김현재 (서울대학교 전기정보공학부) ;
  • 이재구 (서울대학교 전기정보공학부) ;
  • 김규완 (서울대학교 전기정보공학부) ;
  • 윤성로 (서울대학교 전기정보공학부)
  • Received : 2016.10.07
  • Accepted : 2016.12.30
  • Published : 2017.04.15
⟨ Previous Next ⟩

Abstract

Steganalysis is to detect information hidden by steganography inside general data such as images. There are stegoanalysis techniques that use machine learning (ML). Existing ML approaches to steganalysis are based on extracting features from stego images and modeling them. Recently deep learning-based methodologies have shown significant improvements in detection accuracy. However, all the existing methods, including deep learning-based ones, have a critical limitation in that they can only detect stego images that are created by a specific steganography method. In this paper, we propose a generalized steganalysis method that can model multiple types of stego images using deep learning. Through various experiments, we confirm the effectiveness of our approach and envision directions for future research. In particular, we show that our method can detect each type of steganography with the same level of accuracy as that of a steganalysis method dedicated to that type of steganography, thereby demonstrating the general applicability of our approach to multiple types of stego images.

스테그아날리시스(Steganalysis)란 이미지 등 일반적인 자료에 암호화된 정보를 은닉하는 스테가노그래피(Steganography)에 대한 검출 및 분석 방법으로, 기계학습 기반 방법론을 포함한다. 기존 기계학습 기반 스테그아날리시스는 영상(Image)의 특징(Feature) 추출 및 모델링에 기반하며, 최근 딥러닝(Deep Learning)의 적용으로 검출 정확도가 큰 폭으로 향상되었다. 하지만 현존하는 스테그아날리시스 모델은 단일 스테가노그래피 기법에 대해 국한되어 있어 학습에 사용되지 않은 스테고(Stego) 이미지의 경우 검출이 불가능한 결정적 한계를 가진다. 본 연구에서는 다양한 스테가노그래피 기법으로 생성된 스테고 이미지에 딥러닝을 적용하여 스테그아날리시스를 학습하는 범용적 모델을 제안한다. 다양한 실험을 통해 제안 기법의 효용성 및 가능성을 확인하고, 범용적 스테그아날리시스 모델이 각각에 특화된 검출 기법과 유사한 정확도로 스테고 이미지를 검출할 수 있음을 보인다.

Keywords

Acknowledgement

Supported by : 경찰청, 치안과학기술연구개발사업단

References

  1. Huayong, Ge, Huang Mingsheng, and Wang Qian, "Steganography and Steganalysis based on digital image," Image and Signal Processing (CISP), Vol. 1, pp. 252-255, 2011.
  2. T. Pevny, T. Filler, and P. Bas, "Using High-Dimensional Image Models to Perform Highly Undetectable Steganography," International Workshop on Information Hiding, Vol. 6387, No. 1, pp. 161-177, 2010.
  3. V. Holub, J. Fridrich, and T. Denemark, "Universal Distortion Function for Steganography in an Arbitrary Domain," EURASIP Journal on Information Security, pp. 1-13, Jan. 2014.
  4. V. Holub and J. Fridrich, "Designing Steganographic Distortion Using Directional Filters," IEEE International Workshop on Information Forensics and Security (WIFS), pp. 234-239, 2012.
  5. Dhobale, D. D. et al., "Steganography by hiding data in TCP/IP headers," 2010 3rd International Conference on Advanced Computer Theory and Engineering (ICACTE), Vol. 4, pp. 61-65, 2010.
  6. J. Fridrich and J. Kodovsky, "Rich Models for Steganalysis of Digital Images," IEEE Transactions on Information Forensics and Security, Vol. 7, No. 3, pp. 868-882, 2012. https://doi.org/10.1109/TIFS.2012.2190402
  7. J. Kodovsky, J. Fridrich, and V. Holub, "Ensemble Classifiers for Steganalysis of Digital Media," IEEE Transactions on Information Forensics and Security, Vol. 7, No. 2, pp. 432-444, 2012. https://doi.org/10.1109/TIFS.2011.2175919
  8. Yinlong Qian, et al., "Deep Learning for Steganalysis via Convolutional Neural Networks," Proc. of SPIE Media Watermarking, Security, and Forensics, pp. 94090J-94090J-10, 2015.
  9. Lionel Pibre et al., "Deep learning is a good steganalysis tool when embedding key is reused for different images, even if there is a cover source mismatch," Proc. of Media Watermarking, Security, and Forensics, 2016.
  10. LeCun, Yann, et al., "Gradient-based learning applied to document recognition," Proc. of the IEEE, Vol. 86, No. 11, pp. 2278-2324, 1998. https://doi.org/10.1109/5.726791
  11. P. Bas, T. Filler, and T. Pevny, "Break Our Steganographic System': The Ins and Outs of Organizing BOSS," Proc. of the 13th International Conference on Information Hiding, Vol. 6958, pp. 59-70, 2011.