딥러닝 기반 SAR 표적 인식 모델의 적대적 공격에 대한 취약점 분석

과제정보

본 연구는 국방암호기술 특화연구센터(UD210027XD)를 통한 방위사업청과 국방과학연구소의 연구비 지원으로 수행되었습니다

참고문헌

1. Anagnostopoulos, Georgios C. "SVM-based target recognition from synthetic aperture radar images using target region outline descriptors." {Nonlinear Analysis: Theory, Methods \& Applications} 71.12 (2009): e2934-e2939. https://doi.org/10.1016/j.na.2009.07.030
2. Kechagias-Stamatis, Odysseas, and Nabil Aouf. "Automatic target recognition on synthetic aperture radar imagery: A survey." IEEE Aerospace and Electronic Systems Magazine 36.3 (2021): 56-81. https://doi.org/10.1109/MAES.2021.3049857
3. Li, Haifeng, et al. "Adversarial examples for CNN-based SAR image classification: An experience study." IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 14 (2020): 1333-1347. https://doi.org/10.1109/JSTARS.2020.3038683
4. Peng, Bowen, et al. "Scattering model guided adversarial examples for SAR target recognition: Attack and defense." IEEE Transactions on Geoscience and Remote Sensing 60 (2022): 1-17 https://doi.org/10.1109/TGRS.2022.3213305
5. Qin, Weibo, Bo Long, and Feng Wang. "SCMA: A Scattering Center Model Attack on CNN-SAR Target Recognition." {IEEE Geoscience and Remote Sensing Letters 20 (2023): 1-5.
6. Zhou, Junfan, et al. "Attributed scattering center guided adversarial attack for DCNN SAR target recognition." IEEE Geoscience and Remote Sensing Letters 20 (2023): 1-5.
7. Peng, Bowen, et al. "An Empirical Study of Fully Black-Box and Universal Adversarial Attack for SAR Target Recognition." Remote Sensing 14.16 (2022): 4017.
8. hakraborty, Anirban, et al. "Adversarial attacks and defences: A survey." arXiv preprint arXiv:1810.00069 (2018).
9. Su, Jiawei, Danilo Vasconcellos Vargas, and Kouichi Sakurai. "one pixel attack for fooling deep neural networks." IEEE Transactions on Evolutionary Computation 23.5 (2019): 828-841. https://doi.org/10.1109/TEVC.2019.2890858
10. Xiao, Chaowei, et al. "Generating adversarial examples with adversarial networks." arXiv preprint arXiv:1801.02610 (2018).
11. Xie, Cihang, et al. "Improving transferability of adversarial examples with input diversity." Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2019.
12. Wang, Xiaosen, and Kun He. "Enhancing the transferability of adversarial attacks through variance tuning." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021.
13. Malmgren-Hansen, David, and Morten Nobel-J. "Convolutional neural networks for SAR image segmentation." 2015 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT). IEEE, 2015.
14. C. Szegedy, W. Zaremba, I. Sutskever, J. Bruna, D. Erhan, I. Goodfellow, and R. Fergus, "Intriguing Properties of Neural Networks," in ICLR, 2014.
15. Goodfellow, Ian J., Jonathon Shlens, and Christian Szegedy. "Explaining and harnessing adversarial examples." arXiv preprint arXiv:1412.6572 (2014).
16. Moosavi-Dezfooli, Seyed-Mohsen, Alhussein Fawzi, and Pascal Frossard. "Deepfool: a simple and accurate method to fool deep neural networks." Proceedings of the IEEE conference on computer vision and pattern recognition. 2016.
17. A. Madry, A. Makelov, L. Schmidt, D. Tsipras, and A. Vladu, "Towards Deep Learning Models Resistant to Adversarial Attacks," in ICLR, 2018.
18. Carlini, Nicholas, and David Wagner. "Towards evaluating the robustness of neural networks." 2017 ieee symposium on security and privacy (sp). Ieee, 2017.