Improving the robustness of steganalysis in the adversarial environment with Generative Adversarial Network

被引:1
作者
Peng, Ye [1 ]
Yu, Qi [1 ]
Fu, Guobin [1 ]
Zhang, WenWen [2 ]
Duan, ChaoFan [1 ]
机构
[1] Natl Univ Def Technol, Coll Informat & Commun, Wuhan, Peoples R China
[2] Naval Univ Engn, Coll Power Engn, Wuhan, Peoples R China
关键词
Robust steganalysis; Generative Adversarial Network; Adversarial examples; IMAGE; STEGANOGRAPHY; CNN;
D O I
10.1016/j.jisa.2024.103743
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
As technology advances, the steganalysis methods based on Convolutional Neural Network (CNN)can more precisely detect steganography behavior. However, the existing CNN-based steganalysis methods have the problem of insufficient robustness in adversarial environments. Specifically, the steganography methods take advantage of adversarial examples' capacity to deceive deep learning models, adding small perturbations to the stego images to generate adversarial stego images that can successfully evade the detection of CNN-based steganalysis. In this work, we design a robust steganalysis based on the Generative Adversarial Network (GAN), namely RS-GAN. The proposed method reconstructs the suspicious images after learning the distribution of clean images to identify adversarial stego images. Specifically, RS-GAN first disentangles suspicious images to obtain the texture and structure vectors. Afterward, reconstructed images are synthesized based on the above vectors, which have similar distributions to the clean images. In addition, we design an anomaly score calculation module to obtain the reconstruction loss and the discriminator loss and then determine the anomaly scores. Finally, the RS-GAN compares the anomaly scores with the threshold to eliminate the anomaly examples. According to the experimental comparison and analysis, RS-GAN can effectively eliminate the adversarial stego images and increase the robustness of steganalysis in adversarial environments. In the LSUN dataset, when the Fast Gradient Method is employed as the attack with a perturbation value of 0.0013, the accuracy of YeNet decreased to 47.867%, while RS-GAN maintains an accuracy of 72.100%.
引用
收藏
页数:11
相关论文
共 47 条
[1]  
Babaheidarian P, 2019, Decode and transfer: A new steganalysis technique via conditional generative adversarial networks
[2]  
Baluja S, 2017, Adv Neural Inf Process Syst, V2017, P2070
[3]   Deep Residual Network for Steganalysis of Digital Images [J].
Boroumand, Mehdi ;
Chen, Mo ;
Fridrich, Jessica .
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, 2019, 14 (05) :1181-1193
[4]  
Chaumont M, 2020, Deep learning in steganography and steganalysis, DOI [10.1016/B978-0-12-819438-6.00022-0, DOI 10.1016/B978-0-12-819438-6.00022-0]
[5]   Generative Adversarial Networks An overview [J].
Creswell, Antonia ;
White, Tom ;
Dumoulin, Vincent ;
Arulkumaran, Kai ;
Sengupta, Biswa ;
Bharath, Anil A. .
IEEE SIGNAL PROCESSING MAGAZINE, 2018, 35 (01) :53-65
[6]   Rich Models for Steganalysis of Digital Images [J].
Fridrich, Jessica ;
Kodovsky, Jan .
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, 2012, 7 (03) :868-882
[7]   Evasion Attack STeganography: Turning Vulnerability Of Machine Learning To Adversarial Attacks Into A Real-world Application [J].
Ghamizi, Salah ;
Cordy, Maxime ;
Papadakis, Mike ;
Le Traon, Yves .
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW 2021), 2021, :31-40
[8]  
Goodfellow I. J., 2015, INT C LEARN REPR ICL
[9]   Generative Adversarial Networks [J].
Goodfellow, Ian ;
Pouget-Abadie, Jean ;
Mirza, Mehdi ;
Xu, Bing ;
Warde-Farley, David ;
Ozair, Sherjil ;
Courville, Aaron ;
Bengio, Yoshua .
COMMUNICATIONS OF THE ACM, 2020, 63 (11) :139-144
[10]   Towards improving the performance of blind image steganalyzer using third-order SPAM features and ensemble classifier [J].
Hemalatha, J. ;
Sekar, M. ;
Kumar, Chandan ;
Gutub, Adnan ;
Sahu, Aditya Kumar .
JOURNAL OF INFORMATION SECURITY AND APPLICATIONS, 2023, 76