A Lightweight Feature Optimizing Network for Ship Detection in SAR Image
Deep learning-based methods have achieved great success in target detection tasks of computer vision, but when it comes to Synthetic Aperture Radar (SAR) image ship detection, some new challenges appear because of the wide swath of images, diverse appearances of ships and lack of detail information,...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8846704/ |
| _version_ | 1819169335691706368 |
|---|---|
| author | Xiaohan Zhang Haipeng Wang Congan Xu Yafei Lv Chunlong Fu Huachao Xiao You He |
| author_facet | Xiaohan Zhang Haipeng Wang Congan Xu Yafei Lv Chunlong Fu Huachao Xiao You He |
| author_sort | Xiaohan Zhang |
| collection | DOAJ |
| description | Deep learning-based methods have achieved great success in target detection tasks of computer vision, but when it comes to Synthetic Aperture Radar (SAR) image ship detection, some new challenges appear because of the wide swath of images, diverse appearances of ships and lack of detail information, which make the detection inefficient and less effective. Aiming to these issues, in this paper, a lightweight feature optimizing network (LFO-Net) based on popular single shot detector (SSD) model is proposed for single polarization SAR image ship detection. Firstly, a simpler structure called lightweight single shot detector (LSSD) is designed, which can be trained from scratch and can reduce the training and testing time without accuracy cost. Secondly, a new bi-directional feature fusion module including one semantic aggregation block and one feature reuse block is proposed to improve the performance of multi-scale targets detection by enhancing the features of both low feature layers and high feature layers. Then the features are further optimized by leveraging attention mechanism, which is beneficial to catch the silent information more efficiently. A set of experiments are implemented to verify the effectiveness of the proposed method using the public SAR ship detection dataset (SSDD). The results show that the proposed method has significant advantages in both speed and accuracy, and outperforms other state-of-art methods. Additionally, a test on GF-3 satellite SAR data with multiple modes verifies the generalization performance of this model. |
| first_indexed | 2024-12-22T19:17:52Z |
| format | Article |
| id | doaj.art-97ff59fcf8714431970173f6e179f1b2 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T19:17:52Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-97ff59fcf8714431970173f6e179f1b22022-12-21T18:15:29ZengIEEEIEEE Access2169-35362019-01-01714166214167810.1109/ACCESS.2019.29432418846704A Lightweight Feature Optimizing Network for Ship Detection in SAR ImageXiaohan Zhang0https://orcid.org/0000-0002-3693-8194Haipeng Wang1Congan Xu2Yafei Lv3Chunlong Fu4Huachao Xiao5You He6Research Institute of Information Fusion, Naval Aviation University, Yantai, ChinaResearch Institute of Information Fusion, Naval Aviation University, Yantai, ChinaResearch Institute of Information Fusion, Naval Aviation University, Yantai, ChinaResearch Institute of Information Fusion, Naval Aviation University, Yantai, ChinaTroops 90139, PLA, Beijing, ChinaChina Academy of Space Technology, Xi’an, ChinaResearch Institute of Information Fusion, Naval Aviation University, Yantai, ChinaDeep learning-based methods have achieved great success in target detection tasks of computer vision, but when it comes to Synthetic Aperture Radar (SAR) image ship detection, some new challenges appear because of the wide swath of images, diverse appearances of ships and lack of detail information, which make the detection inefficient and less effective. Aiming to these issues, in this paper, a lightweight feature optimizing network (LFO-Net) based on popular single shot detector (SSD) model is proposed for single polarization SAR image ship detection. Firstly, a simpler structure called lightweight single shot detector (LSSD) is designed, which can be trained from scratch and can reduce the training and testing time without accuracy cost. Secondly, a new bi-directional feature fusion module including one semantic aggregation block and one feature reuse block is proposed to improve the performance of multi-scale targets detection by enhancing the features of both low feature layers and high feature layers. Then the features are further optimized by leveraging attention mechanism, which is beneficial to catch the silent information more efficiently. A set of experiments are implemented to verify the effectiveness of the proposed method using the public SAR ship detection dataset (SSDD). The results show that the proposed method has significant advantages in both speed and accuracy, and outperforms other state-of-art methods. Additionally, a test on GF-3 satellite SAR data with multiple modes verifies the generalization performance of this model.https://ieeexplore.ieee.org/document/8846704/SAR ship detectionlightweight modelmulti-scale target detectionbi-directional feature fusionattention mechanism |
| spellingShingle | Xiaohan Zhang Haipeng Wang Congan Xu Yafei Lv Chunlong Fu Huachao Xiao You He A Lightweight Feature Optimizing Network for Ship Detection in SAR Image IEEE Access SAR ship detection lightweight model multi-scale target detection bi-directional feature fusion attention mechanism |
| title | A Lightweight Feature Optimizing Network for Ship Detection in SAR Image |
| title_full | A Lightweight Feature Optimizing Network for Ship Detection in SAR Image |
| title_fullStr | A Lightweight Feature Optimizing Network for Ship Detection in SAR Image |
| title_full_unstemmed | A Lightweight Feature Optimizing Network for Ship Detection in SAR Image |
| title_short | A Lightweight Feature Optimizing Network for Ship Detection in SAR Image |
| title_sort | lightweight feature optimizing network for ship detection in sar image |
| topic | SAR ship detection lightweight model multi-scale target detection bi-directional feature fusion attention mechanism |
| url | https://ieeexplore.ieee.org/document/8846704/ |
| work_keys_str_mv | AT xiaohanzhang alightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT haipengwang alightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT conganxu alightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT yafeilv alightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT chunlongfu alightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT huachaoxiao alightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT youhe alightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT xiaohanzhang lightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT haipengwang lightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT conganxu lightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT yafeilv lightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT chunlongfu lightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT huachaoxiao lightweightfeatureoptimizingnetworkforshipdetectioninsarimage AT youhe lightweightfeatureoptimizingnetworkforshipdetectioninsarimage |