Xiao Gao
Research Subjects
Research Interests
Xiao Gao has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
Di Wang
Xidian University,School of Computer Science and Technology,Xian,China,710071
1 shared publicationKe Li
Xidian University,School of Computer Science and Technology,Xian,China,710071
1 shared publicationPengfei Yang
Xidian University,School of Computer Science and Technology,Xian,China,710071
1 shared publicationWenxuan Zhu
Xidian University,School of Computer Science and Technology,Xian,China,710071
1 shared publication
Latest Publications
- 7 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 6948-6952
Transferable Physical Adversarial Patch Attack for Remote Sensing Object Detection
Deep neural networks (DNNs) have been widely used in remote sensing but demonstrated to be vulnerable with adversarial examples. By adding elaborately designed perturbations on the clean images, DNNs may output wrong prediction. Research on adversarial attack contributes to the study of model robustness. However, previous methods mainly focus on white-box scenario or digital domain for classification tasks, while the vulnerability of remote sensing detectors has not been fully explored. Aiming at attacking black-box remote sensing detectors in physical domain, we propose to generate a transferable physical adversarial patch (TPAP) as the perturbations. Specifically, the initial patch is optimized by a U-Net and modified by the plane mask and position mask before applied to the clean image. By attacking a surrogate model, TPAP can be transferred to the target model. Abundant experimental results validate the attack ability of TPAP and evaluate the robustness of current one-stage detectors.
- 7 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 6948-6952
Transferable Physical Adversarial Patch Attack for Remote Sensing Object Detection
Deep neural networks (DNNs) have been widely used in remote sensing but demonstrated to be vulnerable with adversarial examples. By adding elaborately designed perturbations on the clean images, DNNs may output wrong prediction. Research on adversarial attack contributes to the study of model robustness. However, previous methods mainly focus on white-box scenario or digital domain for classification tasks, while the vulnerability of remote sensing detectors has not been fully explored. Aiming at attacking black-box remote sensing detectors in physical domain, we propose to generate a transferable physical adversarial patch (TPAP) as the perturbations. Specifically, the initial patch is optimized by a U-Net and modified by the plane mask and position mask before applied to the clean image. By attacking a surrogate model, TPAP can be transferred to the target model. Abundant experimental results validate the attack ability of TPAP and evaluate the robustness of current one-stage detectors.