Small Ship Detection of SAR Images Based on Optimized Feature Pyramid and Sample Augmentation
Synthetic aperture radar images have become the latest high-resolution imaging equipment, which can monitor the Earth 24 h a day. More and more deep-learning technologies are applied to ship target detection; however, in complex environments, due to the small target of the ship, problems, such as fa...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10214471/ |
| _version_ | 1797243452033859584 |
|---|---|
| author | Yicheng Gong Zhuo Zhang Jiabao Wen Guipeng Lan Shuai Xiao |
| author_facet | Yicheng Gong Zhuo Zhang Jiabao Wen Guipeng Lan Shuai Xiao |
| author_sort | Yicheng Gong |
| collection | DOAJ |
| description | Synthetic aperture radar images have become the latest high-resolution imaging equipment, which can monitor the Earth 24 h a day. More and more deep-learning technologies are applied to ship target detection; however, in complex environments, due to the small target of the ship, problems, such as false detection and miss detection, often occur. For this reason, SSPNet is proposed with several small-target-augmentation strategies to complete the detection of small ships on the sea. This network is an improvement of FPN. The model uses a context attention module (CAM), scale enhancement module (SEM), and scale selection module (SSM). CAM introduces the attention heat map, SEM uses the residual module to make the network pay more attention to specific scale targets, and SSM introduces deep semantic features into shallow features. A weighted negative sampling strategy is proposed to enable the network to select more representative samples. These modules make the network more suitable for small-target detection. The results on the SSDD dataset show that the model is superior to the existing object detection network, and the average precision (AP<sub>50</sub>) reaches 91.57%. |
| first_indexed | 2024-04-24T18:55:20Z |
| format | Article |
| id | doaj.art-855acb9217ec4f9c9eb814ed7da86920 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-04-24T18:55:20Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-855acb9217ec4f9c9eb814ed7da869202024-03-26T17:35:08ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352023-01-01167385739210.1109/JSTARS.2023.330257510214471Small Ship Detection of SAR Images Based on Optimized Feature Pyramid and Sample AugmentationYicheng Gong0https://orcid.org/0000-0002-4016-6633Zhuo Zhang1https://orcid.org/0000-0002-3946-0720Jiabao Wen2https://orcid.org/0000-0003-2303-9613Guipeng Lan3https://orcid.org/0000-0001-7321-7460Shuai Xiao4https://orcid.org/0000-0003-4058-8120School of Electrical and Information Engineering, Tianjin University, Tianjin, ChinaSchool of Electrical and Information Engineering, Tianjin University, Tianjin, ChinaSchool of Electrical and Information Engineering, Tianjin University, Tianjin, ChinaSchool of Electrical and Information Engineering, Tianjin University, Tianjin, ChinaSchool of Electrical and Information Engineering, Tianjin University, Tianjin, ChinaSynthetic aperture radar images have become the latest high-resolution imaging equipment, which can monitor the Earth 24 h a day. More and more deep-learning technologies are applied to ship target detection; however, in complex environments, due to the small target of the ship, problems, such as false detection and miss detection, often occur. For this reason, SSPNet is proposed with several small-target-augmentation strategies to complete the detection of small ships on the sea. This network is an improvement of FPN. The model uses a context attention module (CAM), scale enhancement module (SEM), and scale selection module (SSM). CAM introduces the attention heat map, SEM uses the residual module to make the network pay more attention to specific scale targets, and SSM introduces deep semantic features into shallow features. A weighted negative sampling strategy is proposed to enable the network to select more representative samples. These modules make the network more suitable for small-target detection. The results on the SSDD dataset show that the model is superior to the existing object detection network, and the average precision (AP<sub>50</sub>) reaches 91.57%.https://ieeexplore.ieee.org/document/10214471/Deep learningobject detectionsample enhancementship detectionsmall-target detectionsynthetic aperture radar (SAR) images |
| spellingShingle | Yicheng Gong Zhuo Zhang Jiabao Wen Guipeng Lan Shuai Xiao Small Ship Detection of SAR Images Based on Optimized Feature Pyramid and Sample Augmentation IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep learning object detection sample enhancement ship detection small-target detection synthetic aperture radar (SAR) images |
| title | Small Ship Detection of SAR Images Based on Optimized Feature Pyramid and Sample Augmentation |
| title_full | Small Ship Detection of SAR Images Based on Optimized Feature Pyramid and Sample Augmentation |
| title_fullStr | Small Ship Detection of SAR Images Based on Optimized Feature Pyramid and Sample Augmentation |
| title_full_unstemmed | Small Ship Detection of SAR Images Based on Optimized Feature Pyramid and Sample Augmentation |
| title_short | Small Ship Detection of SAR Images Based on Optimized Feature Pyramid and Sample Augmentation |
| title_sort | small ship detection of sar images based on optimized feature pyramid and sample augmentation |
| topic | Deep learning object detection sample enhancement ship detection small-target detection synthetic aperture radar (SAR) images |
| url | https://ieeexplore.ieee.org/document/10214471/ |
| work_keys_str_mv | AT yichenggong smallshipdetectionofsarimagesbasedonoptimizedfeaturepyramidandsampleaugmentation AT zhuozhang smallshipdetectionofsarimagesbasedonoptimizedfeaturepyramidandsampleaugmentation AT jiabaowen smallshipdetectionofsarimagesbasedonoptimizedfeaturepyramidandsampleaugmentation AT guipenglan smallshipdetectionofsarimagesbasedonoptimizedfeaturepyramidandsampleaugmentation AT shuaixiao smallshipdetectionofsarimagesbasedonoptimizedfeaturepyramidandsampleaugmentation |