TY - CONF
T1 - FEATURE DECOUPLING BASED ADVERSARIAL EXAMPLES DETECTION METHOD FOR REMOTE SENSING SCENE CLASSIFICATION
AU - Su, Yuru
AU - Mei, Shaohui
AU - Wan, Shuai
N1 - Publisher Copyright:
©2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Deep Neural Networks (DNNs) have demonstrated remarkable effectiveness in remote sensing (RS) image processing. However, they remain vulnerable to adversarial examples, which are generated by adding tiny but purposeful perturbations to clean examples. Such vulnerabilities in critical applications like environmental monitoring and urban planning can lead to significant negative consequences. To mitigate the interference of adversarial examples on DNNs, in this paper, a feature decoupling based adversarial examples detection (FD-AED) method for RS images is proposed, where non-robust features are employed in the detection process. Specifically, a loss function is designed for the decoupler to disentangle the features into robust and non-robust features. Non-robust features are particularly useful because they often contain subtle clues that distinguish between clean and adversarial examples. By focusing on these non-robust features, the adversarial example detector can more effectively capture the differences between clean and adversarial examples. Experimental results indicate that the proposed FD-AED method effectively decouples robust and non-robust features, achieving more precise and reliable detection of adversarial examples.
AB - Deep Neural Networks (DNNs) have demonstrated remarkable effectiveness in remote sensing (RS) image processing. However, they remain vulnerable to adversarial examples, which are generated by adding tiny but purposeful perturbations to clean examples. Such vulnerabilities in critical applications like environmental monitoring and urban planning can lead to significant negative consequences. To mitigate the interference of adversarial examples on DNNs, in this paper, a feature decoupling based adversarial examples detection (FD-AED) method for RS images is proposed, where non-robust features are employed in the detection process. Specifically, a loss function is designed for the decoupler to disentangle the features into robust and non-robust features. Non-robust features are particularly useful because they often contain subtle clues that distinguish between clean and adversarial examples. By focusing on these non-robust features, the adversarial example detector can more effectively capture the differences between clean and adversarial examples. Experimental results indicate that the proposed FD-AED method effectively decouples robust and non-robust features, achieving more precise and reliable detection of adversarial examples.
KW - Adversarial attacks
KW - adversarial examples detection
KW - non-robust features
KW - scene classification
UR - http://www.scopus.com/inward/record.url?scp=85208746001&partnerID=8YFLogxK
U2 - 10.1109/IGARSS53475.2024.10642655
DO - 10.1109/IGARSS53475.2024.10642655
M3 - 论文
AN - SCOPUS:85208746001
SP - 10011
EP - 10014
T2 - 2024 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2024
Y2 - 7 July 2024 through 12 July 2024
ER -