SAR-PATT: A Physical Adversarial Attack for SAR Image Automatic Target Recognition

被引:2
作者
Luo, Binyan [1 ]
Cao, Hang [1 ]
Cui, Jiahao [1 ]
Lv, Xun [1 ]
He, Jinqiang [1 ]
Li, Haifeng [1 ]
Peng, Chengli [1 ]
机构
[1] Cent South Univ, Sch Geosci & Info Phys, South Lushan Rd, Changsha 410083, Peoples R China
基金
中国国家自然科学基金;
关键词
synthetic aperture radar (SAR); deep neural network; adversarial attack; physical attack; SAR simulation; CLASSIFICATION; NETWORK; CNN;
D O I
10.3390/rs17010021
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Deep neural network-based synthetic aperture radar (SAR) automatic target recognition (ATR) systems are susceptible to attack by adversarial examples, which leads to misclassification by the SAR ATR system, resulting in theoretical model robustness problems and security problems in practice. Inspired by optical images, current SAR ATR adversarial example generation is performed in the image domain. However, the imaging principle of SAR images is based on the imaging of the echo signals interacting between the SAR and objects. Generating adversarial examples only in the image domain cannot change the physical world to achieve adversarial attacks. To solve these problems, this article proposes a framework for generating SAR adversarial examples in a 3D physical scene. First, adversarial attacks are implemented in the 2D image space, and the perturbation in the image space is converted into simulated rays that constitute SAR images through backpropagation optimization methods. The mapping between the simulated rays constituting SAR images and the 3D model is established through coordinate transformation, and point correspondence to triangular faces and intensity values to texture parameters are established. Thus, the simulated rays constituting SAR images are mapped to the 3D model, and the perturbation in the 2D image space is converted back to the 3D physical space to obtain the position and intensity of the perturbation in the 3D physical space, thereby achieving physical adversarial attacks. The experimental results show that our attack method can effectively perform SAR adversarial attacks in the physical world. In the digital world, we achieved an average fooling rate of up to 99.02% for three objects in six classification networks. In the physical world, we achieved an average fooling rate of up to 97.87% for these objects, with a certain degree of transferability across the six different network architectures. To the best of our knowledge, this is the first work to implement physical attacks in a full physical simulation condition. Our research establishes a theoretical foundation for the future concealment of SAR targets in practical settings and offers valuable insights for enhancing the attack and defense capabilities of subsequent DNNs in SAR ATR systems.
引用
收藏
页数:19
相关论文
共 61 条
[1]   Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey [J].
Akhtar, Naveed ;
Mian, Ajmal .
IEEE ACCESS, 2018, 6 :14410-14430
[2]  
Auer S., 2011, THESIS TU MUNCHEN
[3]   Ray-Tracing Simulation Techniques for Understanding High-Resolution SAR Images [J].
Auer, Stefan ;
Hinz, Stefan ;
Bamler, Richard .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2010, 48 (03) :1445-1456
[4]   Hybrid GPU-Based Single- and Double-Bounce SAR Simulation [J].
Balz, Timo ;
Stilla, Uwe .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2009, 47 (10) :3519-3529
[5]   SAR SIMULATION OF LOOTING HOLES FOR BETTER IMAGE UNDERSTANDING [J].
Boyogiu, Cem Sonmez ;
Balz, Timo ;
Sultanbekova, Aigerim .
IGARSS 2024-2024 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, IGARSS 2024, 2024, :3337-3340
[6]  
Brent R.P., 2013, Algorithms for minimization without derivatives
[7]  
Brown TB, 2018, Arxiv, DOI arXiv:1712.09665
[8]   Towards Evaluating the Robustness of Neural Networks [J].
Carlini, Nicholas ;
Wagner, David .
2017 IEEE SYMPOSIUM ON SECURITY AND PRIVACY (SP), 2017, :39-57
[9]   An Empirical Study of Adversarial Examples on Remote Sensing Image Scene Classification [J].
Chen, Li ;
Xu, Zewei ;
Li, Qi ;
Peng, Jian ;
Wang, Shaowen ;
Li, Haifeng .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2021, 59 (09) :7419-7433
[10]  
Cui J., 2023, P 2 WORKSH NEW FRONT