Subspace Projection Attention Network for GPR Heterogeneous Clutter Removal
Clutter removal in ground-penetrating radar (GPR) based on deep learning has been studied in recent years. However, existing methods are primarily designed for homogeneous background conditions and utilize only local spatial information via the convolution operation. In order to solve these issues,...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10402071/ |
| _version_ | 1797330370411102208 |
|---|---|
| author | Yanjie Cao Xiaopeng Yang Conglong Guo Dong Li Peng Yin Tian Lan |
| author_facet | Yanjie Cao Xiaopeng Yang Conglong Guo Dong Li Peng Yin Tian Lan |
| author_sort | Yanjie Cao |
| collection | DOAJ |
| description | Clutter removal in ground-penetrating radar (GPR) based on deep learning has been studied in recent years. However, existing methods are primarily designed for homogeneous background conditions and utilize only local spatial information via the convolution operation. In order to solve these issues, a subspace projection attention (SPA) network is proposed for GPR heterogeneous clutter removal in this article. First, a heterogeneous concrete dataset based on a numerical model with randomly placed aggregates is constructed, which incorporates the complex electromagnetic propagation process accurately to improve the effectiveness for heterogeneous clutter removal. In addition, the clutter basis learning neural network is designed by integrating the SPA module into the skip connection paths of U-Net architecture. By learning the subspace basis vectors adaptively, the SPA exploits both local and global spatial information to extract target features precisely. At the same time, the feature maps are projected to the target subspace to remove heterogeneous clutter features. Finally, the performance and effectiveness of proposed method are validated by simulations and experiments. |
| first_indexed | 2024-03-08T07:18:54Z |
| format | Article |
| id | doaj.art-e509e27c524247d2bed801c1caf6c38e |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-03-08T07:18:54Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-e509e27c524247d2bed801c1caf6c38e2024-02-03T00:02:15ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352024-01-01173917392610.1109/JSTARS.2024.335521310402071Subspace Projection Attention Network for GPR Heterogeneous Clutter RemovalYanjie Cao0https://orcid.org/0009-0005-3012-3305Xiaopeng Yang1https://orcid.org/0000-0003-2750-6944Conglong Guo2https://orcid.org/0009-0002-2062-3486Dong Li3https://orcid.org/0000-0003-4766-3808Peng Yin4https://orcid.org/0000-0003-4968-1436Tian Lan5https://orcid.org/0000-0002-2811-2261Beijing Institute of Technology Chongqing Innovation Center, Chongqing, ChinaBeijing Institute of Technology Chongqing Innovation Center, Chongqing, ChinaBeijing Institute of Technology Chongqing Innovation Center, Chongqing, ChinaSchool of Microelectronics and Communication Engineering, Chongqing University, Chongqing, ChinaSchool of Cyber Security, University of Chinese Academy of Sciences, Beijing, ChinaBeijing Institute of Technology Chongqing Innovation Center, Chongqing, ChinaClutter removal in ground-penetrating radar (GPR) based on deep learning has been studied in recent years. However, existing methods are primarily designed for homogeneous background conditions and utilize only local spatial information via the convolution operation. In order to solve these issues, a subspace projection attention (SPA) network is proposed for GPR heterogeneous clutter removal in this article. First, a heterogeneous concrete dataset based on a numerical model with randomly placed aggregates is constructed, which incorporates the complex electromagnetic propagation process accurately to improve the effectiveness for heterogeneous clutter removal. In addition, the clutter basis learning neural network is designed by integrating the SPA module into the skip connection paths of U-Net architecture. By learning the subspace basis vectors adaptively, the SPA exploits both local and global spatial information to extract target features precisely. At the same time, the feature maps are projected to the target subspace to remove heterogeneous clutter features. Finally, the performance and effectiveness of proposed method are validated by simulations and experiments.https://ieeexplore.ieee.org/document/10402071/Clutter removaldeep learningground-penetrating radar (GPR)heterogeneous cluttersubspace projection |
| spellingShingle | Yanjie Cao Xiaopeng Yang Conglong Guo Dong Li Peng Yin Tian Lan Subspace Projection Attention Network for GPR Heterogeneous Clutter Removal IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Clutter removal deep learning ground-penetrating radar (GPR) heterogeneous clutter subspace projection |
| title | Subspace Projection Attention Network for GPR Heterogeneous Clutter Removal |
| title_full | Subspace Projection Attention Network for GPR Heterogeneous Clutter Removal |
| title_fullStr | Subspace Projection Attention Network for GPR Heterogeneous Clutter Removal |
| title_full_unstemmed | Subspace Projection Attention Network for GPR Heterogeneous Clutter Removal |
| title_short | Subspace Projection Attention Network for GPR Heterogeneous Clutter Removal |
| title_sort | subspace projection attention network for gpr heterogeneous clutter removal |
| topic | Clutter removal deep learning ground-penetrating radar (GPR) heterogeneous clutter subspace projection |
| url | https://ieeexplore.ieee.org/document/10402071/ |
| work_keys_str_mv | AT yanjiecao subspaceprojectionattentionnetworkforgprheterogeneousclutterremoval AT xiaopengyang subspaceprojectionattentionnetworkforgprheterogeneousclutterremoval AT conglongguo subspaceprojectionattentionnetworkforgprheterogeneousclutterremoval AT dongli subspaceprojectionattentionnetworkforgprheterogeneousclutterremoval AT pengyin subspaceprojectionattentionnetworkforgprheterogeneousclutterremoval AT tianlan subspaceprojectionattentionnetworkforgprheterogeneousclutterremoval |