Adversarial Patch Attack on Multi-Scale Object Detection for UAV Remote Sensing Images
Although deep learning has received extensive attention and achieved excellent performance in various scenarios, it suffers from adversarial examples to some extent. In particular, physical attack poses a greater threat than digital attack. However, existing research has paid less attention to the p...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/21/5298 |
| _version_ | 1797466679777689600 |
|---|---|
| author | Yichuang Zhang Yu Zhang Jiahao Qi Kangcheng Bin Hao Wen Xunqian Tong Ping Zhong |
| author_facet | Yichuang Zhang Yu Zhang Jiahao Qi Kangcheng Bin Hao Wen Xunqian Tong Ping Zhong |
| author_sort | Yichuang Zhang |
| collection | DOAJ |
| description | Although deep learning has received extensive attention and achieved excellent performance in various scenarios, it suffers from adversarial examples to some extent. In particular, physical attack poses a greater threat than digital attack. However, existing research has paid less attention to the physical attack of object detection in UAV remote sensing images (RSIs). In this work, we carefully analyze the universal adversarial patch attack for multi-scale objects in the field of remote sensing. There are two challenges faced by an adversarial attack in RSIs. On one hand, the number of objects in remote sensing images is more than that of natural images. Therefore, it is difficult for an adversarial patch to show an adversarial effect on all objects when attacking a detector of RSIs. On the other hand, the wide height range of the photography platform causes the size of objects to vary a great deal, which presents challenges for the generation of universal adversarial perturbation for multi-scale objects. To this end, we propose an adversarial attack method of object detection for remote sensing data. One of the key ideas of the proposed method is the novel optimization of the adversarial patch. We aim to attack as many objects as possible by formulating a joint optimization problem. Furthermore, we raise the scale factor to generate a universal adversarial patch that adapts to multi-scale objects, which ensures that the adversarial patch is valid for multi-scale objects in the real world. Extensive experiments demonstrate the superiority of our method against state-of-the-art methods on YOLO-v3 and YOLO-v5. In addition, we also validate the effectiveness of our method in real-world applications. |
| first_indexed | 2024-03-09T18:42:10Z |
| format | Article |
| id | doaj.art-880328a79ada41e99eb99eccb9f82def |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T18:42:10Z |
| publishDate | 2022-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-880328a79ada41e99eb99eccb9f82def2023-11-24T06:36:46ZengMDPI AGRemote Sensing2072-42922022-10-011421529810.3390/rs14215298Adversarial Patch Attack on Multi-Scale Object Detection for UAV Remote Sensing ImagesYichuang Zhang0Yu Zhang1Jiahao Qi2Kangcheng Bin3Hao Wen4Xunqian Tong5Ping Zhong6National Key Laboratory of Science and Technology on Automatic Target Recognition, National University of Defense Technology, Changsha 410073, ChinaNational Key Laboratory of Science and Technology on Automatic Target Recognition, National University of Defense Technology, Changsha 410073, ChinaNational Key Laboratory of Science and Technology on Automatic Target Recognition, National University of Defense Technology, Changsha 410073, ChinaNational Key Laboratory of Science and Technology on Automatic Target Recognition, National University of Defense Technology, Changsha 410073, ChinaNational Key Laboratory of Science and Technology on Automatic Target Recognition, National University of Defense Technology, Changsha 410073, ChinaCollege of Instrumentation and Electrical Engineering, Jilin University, Changchun 130012, ChinaNational Key Laboratory of Science and Technology on Automatic Target Recognition, National University of Defense Technology, Changsha 410073, ChinaAlthough deep learning has received extensive attention and achieved excellent performance in various scenarios, it suffers from adversarial examples to some extent. In particular, physical attack poses a greater threat than digital attack. However, existing research has paid less attention to the physical attack of object detection in UAV remote sensing images (RSIs). In this work, we carefully analyze the universal adversarial patch attack for multi-scale objects in the field of remote sensing. There are two challenges faced by an adversarial attack in RSIs. On one hand, the number of objects in remote sensing images is more than that of natural images. Therefore, it is difficult for an adversarial patch to show an adversarial effect on all objects when attacking a detector of RSIs. On the other hand, the wide height range of the photography platform causes the size of objects to vary a great deal, which presents challenges for the generation of universal adversarial perturbation for multi-scale objects. To this end, we propose an adversarial attack method of object detection for remote sensing data. One of the key ideas of the proposed method is the novel optimization of the adversarial patch. We aim to attack as many objects as possible by formulating a joint optimization problem. Furthermore, we raise the scale factor to generate a universal adversarial patch that adapts to multi-scale objects, which ensures that the adversarial patch is valid for multi-scale objects in the real world. Extensive experiments demonstrate the superiority of our method against state-of-the-art methods on YOLO-v3 and YOLO-v5. In addition, we also validate the effectiveness of our method in real-world applications.https://www.mdpi.com/2072-4292/14/21/5298adversarial examplesremote sensing imagesuniversal adversarial patchobject detectionjoint optimizationscale factor |
| spellingShingle | Yichuang Zhang Yu Zhang Jiahao Qi Kangcheng Bin Hao Wen Xunqian Tong Ping Zhong Adversarial Patch Attack on Multi-Scale Object Detection for UAV Remote Sensing Images Remote Sensing adversarial examples remote sensing images universal adversarial patch object detection joint optimization scale factor |
| title | Adversarial Patch Attack on Multi-Scale Object Detection for UAV Remote Sensing Images |
| title_full | Adversarial Patch Attack on Multi-Scale Object Detection for UAV Remote Sensing Images |
| title_fullStr | Adversarial Patch Attack on Multi-Scale Object Detection for UAV Remote Sensing Images |
| title_full_unstemmed | Adversarial Patch Attack on Multi-Scale Object Detection for UAV Remote Sensing Images |
| title_short | Adversarial Patch Attack on Multi-Scale Object Detection for UAV Remote Sensing Images |
| title_sort | adversarial patch attack on multi scale object detection for uav remote sensing images |
| topic | adversarial examples remote sensing images universal adversarial patch object detection joint optimization scale factor |
| url | https://www.mdpi.com/2072-4292/14/21/5298 |
| work_keys_str_mv | AT yichuangzhang adversarialpatchattackonmultiscaleobjectdetectionforuavremotesensingimages AT yuzhang adversarialpatchattackonmultiscaleobjectdetectionforuavremotesensingimages AT jiahaoqi adversarialpatchattackonmultiscaleobjectdetectionforuavremotesensingimages AT kangchengbin adversarialpatchattackonmultiscaleobjectdetectionforuavremotesensingimages AT haowen adversarialpatchattackonmultiscaleobjectdetectionforuavremotesensingimages AT xunqiantong adversarialpatchattackonmultiscaleobjectdetectionforuavremotesensingimages AT pingzhong adversarialpatchattackonmultiscaleobjectdetectionforuavremotesensingimages |