Boosting Ship Detection in SAR Images With Complementary Pretraining Techniques
Deep learning methods have made significant progress in ship detection in synthetic aperture radar (SAR) images. The pretraining technique is usually adopted to support deep neural networks-based SAR ship detectors due to the scarce labeled SAR images. However, directly leveraging ImageNet pretraini...
| 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/9527093/ |
| _version_ | 1818602158597079040 |
|---|---|
| author | Wei Bao Meiyu Huang Yaqin Zhang Yao Xu Xuejiao Liu Xueshuang Xiang |
| author_facet | Wei Bao Meiyu Huang Yaqin Zhang Yao Xu Xuejiao Liu Xueshuang Xiang |
| author_sort | Wei Bao |
| collection | DOAJ |
| description | Deep learning methods have made significant progress in ship detection in synthetic aperture radar (SAR) images. The pretraining technique is usually adopted to support deep neural networks-based SAR ship detectors due to the scarce labeled SAR images. However, directly leveraging ImageNet pretraining is hard to obtain a good ship detector because of different imaging perspectives and geometry. In this article, to resolve the problem of inconsistent imaging perspectives between ImageNet and earth observations, we propose an optical ship detector (OSD) pretraining technique to transfer the characteristics of ships in earth observations to SAR images from a large-scale aerial image dataset. On the other hand, to handle the problem of different imaging geometry between optical and SAR images, we propose an optical-SAR matching (OSM) pretraining technique, which transfers plentiful texture features from optical images to SAR images by common representation learning on the OSM task. Finally, observing that the OSD pretraining-based SSD has a better recall on sea area while the OSM pretraining-based SSD can reduce false alarms on land area, we combine the predictions of the two detectors through weighted boxes fusion to further improve detection results. Extensive experiments on four SAR ship detection datasets and three representative convolutional neural network-based detection benchmarks are conducted to show the effectiveness and complementarity of the two proposed detectors, and the state-of-the-art performance of the combination of the two detectors. The proposed method won the sixth place of ship detection in SAR images in the 2020 Gaofen challenge. |
| first_indexed | 2024-12-16T13:02:50Z |
| format | Article |
| id | doaj.art-8e6273643a634d63b69a78af52e93469 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-16T13:02:50Z |
| 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-8e6273643a634d63b69a78af52e934692022-12-21T22:30:47ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352021-01-01148941895410.1109/JSTARS.2021.31090029527093Boosting Ship Detection in SAR Images With Complementary Pretraining TechniquesWei Bao0https://orcid.org/0000-0002-2578-1574Meiyu Huang1https://orcid.org/0000-0002-7513-1764Yaqin Zhang2Yao Xu3Xuejiao Liu4https://orcid.org/0000-0002-9331-0934Xueshuang Xiang5https://orcid.org/0000-0001-7794-4876Qian Xuesen Laboratory of Space Technology, China Academy of Space Technology, Beijing, ChinaQian Xuesen Laboratory of Space Technology, China Academy of Space Technology, Beijing, ChinaQian Xuesen Laboratory of Space Technology, China Academy of Space Technology, Beijing, ChinaQian Xuesen Laboratory of Space Technology, China Academy of Space Technology, Beijing, ChinaQian Xuesen Laboratory of Space Technology, China Academy of Space Technology, Beijing, ChinaQian Xuesen Laboratory of Space Technology, China Academy of Space Technology, Beijing, ChinaDeep learning methods have made significant progress in ship detection in synthetic aperture radar (SAR) images. The pretraining technique is usually adopted to support deep neural networks-based SAR ship detectors due to the scarce labeled SAR images. However, directly leveraging ImageNet pretraining is hard to obtain a good ship detector because of different imaging perspectives and geometry. In this article, to resolve the problem of inconsistent imaging perspectives between ImageNet and earth observations, we propose an optical ship detector (OSD) pretraining technique to transfer the characteristics of ships in earth observations to SAR images from a large-scale aerial image dataset. On the other hand, to handle the problem of different imaging geometry between optical and SAR images, we propose an optical-SAR matching (OSM) pretraining technique, which transfers plentiful texture features from optical images to SAR images by common representation learning on the OSM task. Finally, observing that the OSD pretraining-based SSD has a better recall on sea area while the OSM pretraining-based SSD can reduce false alarms on land area, we combine the predictions of the two detectors through weighted boxes fusion to further improve detection results. Extensive experiments on four SAR ship detection datasets and three representative convolutional neural network-based detection benchmarks are conducted to show the effectiveness and complementarity of the two proposed detectors, and the state-of-the-art performance of the combination of the two detectors. The proposed method won the sixth place of ship detection in SAR images in the 2020 Gaofen challenge.https://ieeexplore.ieee.org/document/9527093/Common representation learningoptical ship detector (OSD) pretrainingoptical-SAR matching (OSM) pretrainingship detectionweighted boxes fusion (WBF) |
| spellingShingle | Wei Bao Meiyu Huang Yaqin Zhang Yao Xu Xuejiao Liu Xueshuang Xiang Boosting Ship Detection in SAR Images With Complementary Pretraining Techniques IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Common representation learning optical ship detector (OSD) pretraining optical-SAR matching (OSM) pretraining ship detection weighted boxes fusion (WBF) |
| title | Boosting Ship Detection in SAR Images With Complementary Pretraining Techniques |
| title_full | Boosting Ship Detection in SAR Images With Complementary Pretraining Techniques |
| title_fullStr | Boosting Ship Detection in SAR Images With Complementary Pretraining Techniques |
| title_full_unstemmed | Boosting Ship Detection in SAR Images With Complementary Pretraining Techniques |
| title_short | Boosting Ship Detection in SAR Images With Complementary Pretraining Techniques |
| title_sort | boosting ship detection in sar images with complementary pretraining techniques |
| topic | Common representation learning optical ship detector (OSD) pretraining optical-SAR matching (OSM) pretraining ship detection weighted boxes fusion (WBF) |
| url | https://ieeexplore.ieee.org/document/9527093/ |
| work_keys_str_mv | AT weibao boostingshipdetectioninsarimageswithcomplementarypretrainingtechniques AT meiyuhuang boostingshipdetectioninsarimageswithcomplementarypretrainingtechniques AT yaqinzhang boostingshipdetectioninsarimageswithcomplementarypretrainingtechniques AT yaoxu boostingshipdetectioninsarimageswithcomplementarypretrainingtechniques AT xuejiaoliu boostingshipdetectioninsarimageswithcomplementarypretrainingtechniques AT xueshuangxiang boostingshipdetectioninsarimageswithcomplementarypretrainingtechniques |