Light Attack: A Physical World Real-Time Attack Against Object Classifiers
It is well known that deep neural networks (DNNs) are vulnerable to adversarial examples. In the digital world, most of the existing work makes classifiers or detectors fail by adding perturbations that are imperceptible to humans. In the physical world, existing work mostly invalidates classifiers...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9791340/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849708248197758976 |
|---|---|
| author | Ruizhe Hu Ting Rui Yan Ouyang Jinkang Wang Qunyan Jiang Yinan Du |
| author_facet | Ruizhe Hu Ting Rui Yan Ouyang Jinkang Wang Qunyan Jiang Yinan Du |
| author_sort | Ruizhe Hu |
| collection | DOAJ |
| description | It is well known that deep neural networks (DNNs) are vulnerable to adversarial examples. In the digital world, most of the existing work makes classifiers or detectors fail by adding perturbations that are imperceptible to humans. In the physical world, existing work mostly invalidates classifiers or detectors by adding large unrealistic perturbations. In this paper, we attack the target classifiers from a new perspective of light attack. First, we define three lighting modes (Line, Point, Area), then we control the wavelength, intensity and position of light to accomplish adversarial attack, and finally we conduct ablation experiments and compare our method with existing mainstream methods. Experiments demonstrate that our proposed method is effective in both digital and physical worlds, and in darker environments, our attack method is much better than existing ones. Light attack enriches the current family of the adversarial attack while paving the way for future defense against light attacks as well as the study of light attack laws in nature. |
| format | Article |
| id | doaj-art-5ea9f9bb2d8e481a84f8dd054ec7e23f |
| institution | DOAJ |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-5ea9f9bb2d8e481a84f8dd054ec7e23f2025-08-20T03:15:43ZengIEEEIEEE Access2169-35362025-01-0113366013661010.1109/ACCESS.2022.31811979791340Light Attack: A Physical World Real-Time Attack Against Object ClassifiersRuizhe Hu0https://orcid.org/0000-0002-4891-9820Ting Rui1https://orcid.org/0000-0002-2949-5874Yan Ouyang2https://orcid.org/0000-0002-4481-9662Jinkang Wang3https://orcid.org/0000-0001-8866-6744Qunyan Jiang4Yinan Du5Department of Mechanical Engineering, College of Field Engineering, Army Engineering University of PLA, Nanjing, ChinaDepartment of Mechanical Engineering, College of Field Engineering, Army Engineering University of PLA, Nanjing, ChinaDepartment of Mechanical Engineering, College of Field Engineering, Army Engineering University of PLA, Nanjing, ChinaDepartment of Mechanical Engineering, College of Field Engineering, Army Engineering University of PLA, Nanjing, ChinaDepartment of Mechanical Engineering, College of Field Engineering, Army Engineering University of PLA, Nanjing, ChinaDepartment of Mechanical Engineering, College of Field Engineering, Army Engineering University of PLA, Nanjing, ChinaIt is well known that deep neural networks (DNNs) are vulnerable to adversarial examples. In the digital world, most of the existing work makes classifiers or detectors fail by adding perturbations that are imperceptible to humans. In the physical world, existing work mostly invalidates classifiers or detectors by adding large unrealistic perturbations. In this paper, we attack the target classifiers from a new perspective of light attack. First, we define three lighting modes (Line, Point, Area), then we control the wavelength, intensity and position of light to accomplish adversarial attack, and finally we conduct ablation experiments and compare our method with existing mainstream methods. Experiments demonstrate that our proposed method is effective in both digital and physical worlds, and in darker environments, our attack method is much better than existing ones. Light attack enriches the current family of the adversarial attack while paving the way for future defense against light attacks as well as the study of light attack laws in nature.https://ieeexplore.ieee.org/document/9791340/Adversarial exampleadversarial attacklight attackreal-time attackobject classifier |
| spellingShingle | Ruizhe Hu Ting Rui Yan Ouyang Jinkang Wang Qunyan Jiang Yinan Du Light Attack: A Physical World Real-Time Attack Against Object Classifiers IEEE Access Adversarial example adversarial attack light attack real-time attack object classifier |
| title | Light Attack: A Physical World Real-Time Attack Against Object Classifiers |
| title_full | Light Attack: A Physical World Real-Time Attack Against Object Classifiers |
| title_fullStr | Light Attack: A Physical World Real-Time Attack Against Object Classifiers |
| title_full_unstemmed | Light Attack: A Physical World Real-Time Attack Against Object Classifiers |
| title_short | Light Attack: A Physical World Real-Time Attack Against Object Classifiers |
| title_sort | light attack a physical world real time attack against object classifiers |
| topic | Adversarial example adversarial attack light attack real-time attack object classifier |
| url | https://ieeexplore.ieee.org/document/9791340/ |
| work_keys_str_mv | AT ruizhehu lightattackaphysicalworldrealtimeattackagainstobjectclassifiers AT tingrui lightattackaphysicalworldrealtimeattackagainstobjectclassifiers AT yanouyang lightattackaphysicalworldrealtimeattackagainstobjectclassifiers AT jinkangwang lightattackaphysicalworldrealtimeattackagainstobjectclassifiers AT qunyanjiang lightattackaphysicalworldrealtimeattackagainstobjectclassifiers AT yinandu lightattackaphysicalworldrealtimeattackagainstobjectclassifiers |