Adversarial Examples for CNN-Based SAR Image Classification: An Experience Study
Synthetic aperture radar (SAR) has all-day and all-weather characteristics and plays an extremely important role in the military field. The breakthroughs in deep learning methods represented by convolutional neural network (CNN) models have greatly improved the SAR image recognition accuracy. Classi...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9261933/ |
| _version_ | 1818821073903288320 |
|---|---|
| author | Haifeng Li Haikuo Huang Li Chen Jian Peng Haozhe Huang Zhenqi Cui Xiaoming Mei Guohua Wu |
| author_facet | Haifeng Li Haikuo Huang Li Chen Jian Peng Haozhe Huang Zhenqi Cui Xiaoming Mei Guohua Wu |
| author_sort | Haifeng Li |
| collection | DOAJ |
| description | Synthetic aperture radar (SAR) has all-day and all-weather characteristics and plays an extremely important role in the military field. The breakthroughs in deep learning methods represented by convolutional neural network (CNN) models have greatly improved the SAR image recognition accuracy. Classification models based on CNNs can perform high-precision classification, but there are security problems against adversarial examples (AEs). However, the research on AEs is mostly limited to natural images, and remote sensing images (SAR, multispectral, etc.) have not been extensively studied. To explore the basic characteristics of AEs of SAR images (ASIs), we use two classic white-box attack methods to generate ASIs from two SAR image classification datasets and then evaluate the vulnerability of six commonly used CNNs. The results show that ASIs are quite effective in fooling CNNs trained on SAR images, as indicated by the obtained high attack success rate. Due to the structural differences among CNNs, different CNNs present different vulnerabilities in the face of ASIs. We found that ASIs generated by nontarget attack algorithms feature attack selectivity, which is related to the feature space distribution of the original SAR images and the decision boundary of the classification model. We propose the sample-boundary-based AE selectivity distance to successfully explain the attack selectivity of ASIs. We also analyze the effects of image parameters, such as image size and number of channels, on the attack success rate of ASIs through parameter sensitivity. The experimental results of this study provide data support and an effective reference for attacks on and the defense capabilities of various CNNs with regard to AEs in SAR image classification models. |
| first_indexed | 2024-12-18T23:02:24Z |
| format | Article |
| id | doaj.art-ebc82d05b5df4fe782f32cefed503373 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-18T23:02:24Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-ebc82d05b5df4fe782f32cefed5033732022-12-21T20:48:33ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352021-01-01141333134710.1109/JSTARS.2020.30386839261933Adversarial Examples for CNN-Based SAR Image Classification: An Experience StudyHaifeng Li0https://orcid.org/0000-0003-1173-6593Haikuo Huang1Li Chen2Jian Peng3https://orcid.org/0000-0002-1820-4015Haozhe Huang4Zhenqi Cui5Xiaoming Mei6Guohua Wu7https://orcid.org/0000-0003-1552-9620School of Geosciences and Info-Physics, Central South University, Changsha, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha, ChinaSchool of Traffic and Transportation Engineering, Central South University, Changsha, ChinaSynthetic aperture radar (SAR) has all-day and all-weather characteristics and plays an extremely important role in the military field. The breakthroughs in deep learning methods represented by convolutional neural network (CNN) models have greatly improved the SAR image recognition accuracy. Classification models based on CNNs can perform high-precision classification, but there are security problems against adversarial examples (AEs). However, the research on AEs is mostly limited to natural images, and remote sensing images (SAR, multispectral, etc.) have not been extensively studied. To explore the basic characteristics of AEs of SAR images (ASIs), we use two classic white-box attack methods to generate ASIs from two SAR image classification datasets and then evaluate the vulnerability of six commonly used CNNs. The results show that ASIs are quite effective in fooling CNNs trained on SAR images, as indicated by the obtained high attack success rate. Due to the structural differences among CNNs, different CNNs present different vulnerabilities in the face of ASIs. We found that ASIs generated by nontarget attack algorithms feature attack selectivity, which is related to the feature space distribution of the original SAR images and the decision boundary of the classification model. We propose the sample-boundary-based AE selectivity distance to successfully explain the attack selectivity of ASIs. We also analyze the effects of image parameters, such as image size and number of channels, on the attack success rate of ASIs through parameter sensitivity. The experimental results of this study provide data support and an effective reference for attacks on and the defense capabilities of various CNNs with regard to AEs in SAR image classification models.https://ieeexplore.ieee.org/document/9261933/Adversarial example (AE)convolutional neural network (CNN)synthetic aperture radar (SAR) |
| spellingShingle | Haifeng Li Haikuo Huang Li Chen Jian Peng Haozhe Huang Zhenqi Cui Xiaoming Mei Guohua Wu Adversarial Examples for CNN-Based SAR Image Classification: An Experience Study IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Adversarial example (AE) convolutional neural network (CNN) synthetic aperture radar (SAR) |
| title | Adversarial Examples for CNN-Based SAR Image Classification: An Experience Study |
| title_full | Adversarial Examples for CNN-Based SAR Image Classification: An Experience Study |
| title_fullStr | Adversarial Examples for CNN-Based SAR Image Classification: An Experience Study |
| title_full_unstemmed | Adversarial Examples for CNN-Based SAR Image Classification: An Experience Study |
| title_short | Adversarial Examples for CNN-Based SAR Image Classification: An Experience Study |
| title_sort | adversarial examples for cnn based sar image classification an experience study |
| topic | Adversarial example (AE) convolutional neural network (CNN) synthetic aperture radar (SAR) |
| url | https://ieeexplore.ieee.org/document/9261933/ |
| work_keys_str_mv | AT haifengli adversarialexamplesforcnnbasedsarimageclassificationanexperiencestudy AT haikuohuang adversarialexamplesforcnnbasedsarimageclassificationanexperiencestudy AT lichen adversarialexamplesforcnnbasedsarimageclassificationanexperiencestudy AT jianpeng adversarialexamplesforcnnbasedsarimageclassificationanexperiencestudy AT haozhehuang adversarialexamplesforcnnbasedsarimageclassificationanexperiencestudy AT zhenqicui adversarialexamplesforcnnbasedsarimageclassificationanexperiencestudy AT xiaomingmei adversarialexamplesforcnnbasedsarimageclassificationanexperiencestudy AT guohuawu adversarialexamplesforcnnbasedsarimageclassificationanexperiencestudy |