HOFA-Net: A High-Order Feature Association Network for Dense Object Detection in Remote Sensing
In the remote sensing field, deep learning-based methods have become mainstream for remote sensing image object detection in recent years. However, traditional methods, such as convolutional neural networks (CNNs), mainly ignore the dependencies between features, failing to capture the spatial relat...
| 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/10330558/ |
| _version_ | 1827399034302627840 |
|---|---|
| author | YunPeng Xu Xin Wu Li Wang Lianming Xu Zhengyu Shao Aiguo Fei |
| author_facet | YunPeng Xu Xin Wu Li Wang Lianming Xu Zhengyu Shao Aiguo Fei |
| author_sort | YunPeng Xu |
| collection | DOAJ |
| description | In the remote sensing field, deep learning-based methods have become mainstream for remote sensing image object detection in recent years. However, traditional methods, such as convolutional neural networks (CNNs), mainly ignore the dependencies between features, failing to capture the spatial relationships and relative positions of objects, which affects the detection performance of dense objects, especially small-size objects. To this end, a high-order feature association network (HOFA-Net) for dense object detection in remote sensing has been proposed to better capture the interdependencies between features of channel and spatial dimensions, yielding more distinguishable features. First, we employ CNNs to learn high-level but low-resolution features of objects. To capture feature interdependencies while retaining crucial information, we design a feature association module based on size adaptation nonlocal. This module partitions the low-resolution and high-level features into local regions and utilizes nonlocal residual connections to capture the local contextual information of objects. In addition, we introduce a high-order feature association (HFA) module designed to learn nonlinear feature correlations and interdependencies within the features. In addition, a covariance normalization acceleration strategy is introduced to accelerate computation. Experimental results on two public remote sensing datasets, including the DOTA dataset and the Tiny Person dataset, demonstrate the superiority and effectiveness of the proposed method through comparative experiments. |
| first_indexed | 2024-03-08T19:38:42Z |
| format | Article |
| id | doaj.art-a892155424c447abb1bba47bc3c8216c |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-03-08T19:38:42Z |
| 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-a892155424c447abb1bba47bc3c8216c2023-12-26T00:00:46ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352024-01-01171513152210.1109/JSTARS.2023.333528810330558HOFA-Net: A High-Order Feature Association Network for Dense Object Detection in Remote SensingYunPeng Xu0Xin Wu1https://orcid.org/0000-0002-1733-3560Li Wang2https://orcid.org/0000-0002-0973-1614Lianming Xu3Zhengyu Shao4https://orcid.org/0009-0002-3598-7610Aiguo Fei5School of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing, ChinaSchool of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing, ChinaSchool of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing, ChinaSchool of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing, ChinaSchool of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing, ChinaSchool of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing, ChinaIn the remote sensing field, deep learning-based methods have become mainstream for remote sensing image object detection in recent years. However, traditional methods, such as convolutional neural networks (CNNs), mainly ignore the dependencies between features, failing to capture the spatial relationships and relative positions of objects, which affects the detection performance of dense objects, especially small-size objects. To this end, a high-order feature association network (HOFA-Net) for dense object detection in remote sensing has been proposed to better capture the interdependencies between features of channel and spatial dimensions, yielding more distinguishable features. First, we employ CNNs to learn high-level but low-resolution features of objects. To capture feature interdependencies while retaining crucial information, we design a feature association module based on size adaptation nonlocal. This module partitions the low-resolution and high-level features into local regions and utilizes nonlocal residual connections to capture the local contextual information of objects. In addition, we introduce a high-order feature association (HFA) module designed to learn nonlinear feature correlations and interdependencies within the features. In addition, a covariance normalization acceleration strategy is introduced to accelerate computation. Experimental results on two public remote sensing datasets, including the DOTA dataset and the Tiny Person dataset, demonstrate the superiority and effectiveness of the proposed method through comparative experiments.https://ieeexplore.ieee.org/document/10330558/Convolutional neural networks (CNNs)covariance normalizationhigh-order feature associationobject detectionremote sensing |
| spellingShingle | YunPeng Xu Xin Wu Li Wang Lianming Xu Zhengyu Shao Aiguo Fei HOFA-Net: A High-Order Feature Association Network for Dense Object Detection in Remote Sensing IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Convolutional neural networks (CNNs) covariance normalization high-order feature association object detection remote sensing |
| title | HOFA-Net: A High-Order Feature Association Network for Dense Object Detection in Remote Sensing |
| title_full | HOFA-Net: A High-Order Feature Association Network for Dense Object Detection in Remote Sensing |
| title_fullStr | HOFA-Net: A High-Order Feature Association Network for Dense Object Detection in Remote Sensing |
| title_full_unstemmed | HOFA-Net: A High-Order Feature Association Network for Dense Object Detection in Remote Sensing |
| title_short | HOFA-Net: A High-Order Feature Association Network for Dense Object Detection in Remote Sensing |
| title_sort | hofa net a high order feature association network for dense object detection in remote sensing |
| topic | Convolutional neural networks (CNNs) covariance normalization high-order feature association object detection remote sensing |
| url | https://ieeexplore.ieee.org/document/10330558/ |
| work_keys_str_mv | AT yunpengxu hofanetahighorderfeatureassociationnetworkfordenseobjectdetectioninremotesensing AT xinwu hofanetahighorderfeatureassociationnetworkfordenseobjectdetectioninremotesensing AT liwang hofanetahighorderfeatureassociationnetworkfordenseobjectdetectioninremotesensing AT lianmingxu hofanetahighorderfeatureassociationnetworkfordenseobjectdetectioninremotesensing AT zhengyushao hofanetahighorderfeatureassociationnetworkfordenseobjectdetectioninremotesensing AT aiguofei hofanetahighorderfeatureassociationnetworkfordenseobjectdetectioninremotesensing |