Multilevel Pyramid Feature Extraction and Task Decoupling Network for SAR Ship Detection
Synthetic aperture radar (SAR) target detection plays a crucial role in both military and civilian fields, attracting significant attention from researchers globally. CenterNet, a single-stage target detection method, is known for its high detection speed and accuracy by eliminating anchor-related c...
| 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/10403783/ |
| _version_ | 1827370990369243136 |
|---|---|
| author | Yanshan Li Wenjun Liu Ruo Qi |
| author_facet | Yanshan Li Wenjun Liu Ruo Qi |
| author_sort | Yanshan Li |
| collection | DOAJ |
| description | Synthetic aperture radar (SAR) target detection plays a crucial role in both military and civilian fields, attracting significant attention from researchers globally. CenterNet, a single-stage target detection method, is known for its high detection speed and accuracy by eliminating anchor-related calculations and nonmaximum suppression. However, directly applying CenterNet to SAR ship detection poses challenges due to the distinctive characteristics of SAR images, including lower resolution, lower signal-to-noise ratio, and larger ship aspect ratios. To address these challenges, we propose MPDNet. which introduces a multilevel pyramid feature extraction module (MP-FEM) to replace the encoding–decoding structure in CenterNet. MP-FEM employs multilevel pyramid and channel compression to fuse multiscale SAR image features and acquire deep features quickly. Second, we propose the convolution channel attention module, which improves the multilayer perceptron in the common pooling attention mechanism into a multistage and 1-D convolution. Therefore, the feature extraction capability of MP-FEM is further refined. Furthermore, we propose the detection task decoupling module (DTDM), which considers the characteristics of SAR ships and effectively detects smaller targets of different sizes, distinguishing the centers and sizes of densely arranged ships. DTDM extracts task-related features from the original feature map before inputting it into the three detection headers, thereby addressing the problem of task coupling in CenterNet's detection header module for SAR ship detection. Finally, the experimental results on SSDD dataset and SAR-ship-dataset show that the proposed network can significantly improve the SAR target detection accuracy. |
| first_indexed | 2024-03-08T10:31:39Z |
| format | Article |
| id | doaj.art-9c0f8f33a3d44c65ac720f5210c25ab5 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-03-08T10:31:39Z |
| 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-9c0f8f33a3d44c65ac720f5210c25ab52024-01-27T00:00:53ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352024-01-01173560357010.1109/JSTARS.2023.334745410403783Multilevel Pyramid Feature Extraction and Task Decoupling Network for SAR Ship DetectionYanshan Li0https://orcid.org/0000-0002-8814-4628Wenjun Liu1https://orcid.org/0009-0003-2746-8201Ruo Qi2https://orcid.org/0009-0008-6643-5089ATR National Key Laboratory of Defense Technology, Shenzhen University, Shenzhen, ChinaATR National Key Laboratory of Defense Technology, Shenzhen University, Shenzhen, ChinaATR National Key Laboratory of Defense Technology, Shenzhen University, Shenzhen, ChinaSynthetic aperture radar (SAR) target detection plays a crucial role in both military and civilian fields, attracting significant attention from researchers globally. CenterNet, a single-stage target detection method, is known for its high detection speed and accuracy by eliminating anchor-related calculations and nonmaximum suppression. However, directly applying CenterNet to SAR ship detection poses challenges due to the distinctive characteristics of SAR images, including lower resolution, lower signal-to-noise ratio, and larger ship aspect ratios. To address these challenges, we propose MPDNet. which introduces a multilevel pyramid feature extraction module (MP-FEM) to replace the encoding–decoding structure in CenterNet. MP-FEM employs multilevel pyramid and channel compression to fuse multiscale SAR image features and acquire deep features quickly. Second, we propose the convolution channel attention module, which improves the multilayer perceptron in the common pooling attention mechanism into a multistage and 1-D convolution. Therefore, the feature extraction capability of MP-FEM is further refined. Furthermore, we propose the detection task decoupling module (DTDM), which considers the characteristics of SAR ships and effectively detects smaller targets of different sizes, distinguishing the centers and sizes of densely arranged ships. DTDM extracts task-related features from the original feature map before inputting it into the three detection headers, thereby addressing the problem of task coupling in CenterNet's detection header module for SAR ship detection. Finally, the experimental results on SSDD dataset and SAR-ship-dataset show that the proposed network can significantly improve the SAR target detection accuracy.https://ieeexplore.ieee.org/document/10403783/CenterNetmultilevel feature pyramidsynthetic aperture radar (SAR) imagetarget detection |
| spellingShingle | Yanshan Li Wenjun Liu Ruo Qi Multilevel Pyramid Feature Extraction and Task Decoupling Network for SAR Ship Detection IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing CenterNet multilevel feature pyramid synthetic aperture radar (SAR) image target detection |
| title | Multilevel Pyramid Feature Extraction and Task Decoupling Network for SAR Ship Detection |
| title_full | Multilevel Pyramid Feature Extraction and Task Decoupling Network for SAR Ship Detection |
| title_fullStr | Multilevel Pyramid Feature Extraction and Task Decoupling Network for SAR Ship Detection |
| title_full_unstemmed | Multilevel Pyramid Feature Extraction and Task Decoupling Network for SAR Ship Detection |
| title_short | Multilevel Pyramid Feature Extraction and Task Decoupling Network for SAR Ship Detection |
| title_sort | multilevel pyramid feature extraction and task decoupling network for sar ship detection |
| topic | CenterNet multilevel feature pyramid synthetic aperture radar (SAR) image target detection |
| url | https://ieeexplore.ieee.org/document/10403783/ |
| work_keys_str_mv | AT yanshanli multilevelpyramidfeatureextractionandtaskdecouplingnetworkforsarshipdetection AT wenjunliu multilevelpyramidfeatureextractionandtaskdecouplingnetworkforsarshipdetection AT ruoqi multilevelpyramidfeatureextractionandtaskdecouplingnetworkforsarshipdetection |