Adversarial Robustness Evaluation of Deep Convolutional Neural Network Based SAR ATR Algorithm

Robustness, both to accident and to malevolent perturbations, is a crucial determinant of the successful deployment of deep convolutional neural network based SAR ATR systems in various security-sensitive applications. This paper performs a detailed adversarial robustness evaluation of deep convolut...

Full description

Saved in:
Bibliographic Details
Published inIEEE International Geoscience and Remote Sensing Symposium proceedings pp. 5263 - 5266
Main Authors Sun, Hao, Xu, Yanjie, Kuang, Gangyao, Chen, Jin
Format Conference Proceeding
LanguageEnglish
Published IEEE 11.07.2021
Subjects
adversarial robustness
Convolutional neural networks
deep learning
Distance measurement
Distortion
feature attribution
Geoscience and remote sensing
Perturbation methods
Robustness
SAR
Target recognition
Online AccessGet full text
ISSN2153-7003
DOI10.1109/IGARSS47720.2021.9554783

Cover

More Information
Summary:Robustness, both to accident and to malevolent perturbations, is a crucial determinant of the successful deployment of deep convolutional neural network based SAR ATR systems in various security-sensitive applications. This paper performs a detailed adversarial robustness evaluation of deep convolutional neural network based SAR ATR models across two public available SAR target recognition datasets. For each model, seven different adversarial perturbations, ranging from gradient based optimization to self-supervised feature distortion, are generated for each testing image. Besides adversarial average recognition accuracy, feature attribution techniques have also been adopted to analyze the feature diffusion effect of adversarial attacks, which promotes the understanding of vulnerability of deep learning models.
ISSN:2153-7003
DOI:10.1109/IGARSS47720.2021.9554783