A Multi-Scale Feature Pyramid Network for Detection and Instance Segmentation of Marine Ships in SAR Images
In the remote sensing field, synthetic aperture radar (SAR) is a type of active microwave imaging sensor working in all-weather and all-day conditions, providing high-resolution SAR images of objects such as marine ships. Detection and instance segmentation of marine ships in SAR images has become a...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/24/6312 |
| _version_ | 1797455416870830080 |
|---|---|
| author | Zequn Sun Chunning Meng Jierong Cheng Zhiqing Zhang Shengjiang Chang |
| author_facet | Zequn Sun Chunning Meng Jierong Cheng Zhiqing Zhang Shengjiang Chang |
| author_sort | Zequn Sun |
| collection | DOAJ |
| description | In the remote sensing field, synthetic aperture radar (SAR) is a type of active microwave imaging sensor working in all-weather and all-day conditions, providing high-resolution SAR images of objects such as marine ships. Detection and instance segmentation of marine ships in SAR images has become an important question in remote sensing, but current deep learning models cannot accurately quantify marine ships because of the multi-scale property of marine ships in SAR images. In this paper, we propose a multi-scale feature pyramid network (MS-FPN) to achieve the simultaneous detection and instance segmentation of marine ships in SAR images. The proposed MS-FPN model uses a pyramid structure, and it is mainly composed of two proposed modules, namely the atrous convolutional pyramid (ACP) module and the multi-scale attention mechanism (MSAM) module. The ACP module is designed to extract both the shallow and deep feature maps, and these multi-scale feature maps are crucial for the description of multi-scale marine ships, especially the small ones. The MSAM module is designed to adaptively learn and select important feature maps obtained from different scales, leading to improved detection and segmentation accuracy. Quantitative comparison of the proposed MS-FPN model with several classical and recently developed deep learning models, using the high-resolution SAR images dataset (HRSID) that contains multi-scale marine ship SAR images, demonstrated the superior performance of MS-FPN over other models. |
| first_indexed | 2024-03-09T15:54:06Z |
| format | Article |
| id | doaj.art-0323030ce8964fa191b0d721744e1527 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T15:54:06Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-0323030ce8964fa191b0d721744e15272023-11-24T17:47:36ZengMDPI AGRemote Sensing2072-42922022-12-011424631210.3390/rs14246312A Multi-Scale Feature Pyramid Network for Detection and Instance Segmentation of Marine Ships in SAR ImagesZequn Sun0Chunning Meng1Jierong Cheng2Zhiqing Zhang3Shengjiang Chang4Institute of Modern Optics, Nankai University, Tianjin 300350, ChinaChina Coast Guard Academy, Ningbo 315801, ChinaInstitute of Modern Optics, Nankai University, Tianjin 300350, ChinaInstitute of Modern Optics, Nankai University, Tianjin 300350, ChinaInstitute of Modern Optics, Nankai University, Tianjin 300350, ChinaIn the remote sensing field, synthetic aperture radar (SAR) is a type of active microwave imaging sensor working in all-weather and all-day conditions, providing high-resolution SAR images of objects such as marine ships. Detection and instance segmentation of marine ships in SAR images has become an important question in remote sensing, but current deep learning models cannot accurately quantify marine ships because of the multi-scale property of marine ships in SAR images. In this paper, we propose a multi-scale feature pyramid network (MS-FPN) to achieve the simultaneous detection and instance segmentation of marine ships in SAR images. The proposed MS-FPN model uses a pyramid structure, and it is mainly composed of two proposed modules, namely the atrous convolutional pyramid (ACP) module and the multi-scale attention mechanism (MSAM) module. The ACP module is designed to extract both the shallow and deep feature maps, and these multi-scale feature maps are crucial for the description of multi-scale marine ships, especially the small ones. The MSAM module is designed to adaptively learn and select important feature maps obtained from different scales, leading to improved detection and segmentation accuracy. Quantitative comparison of the proposed MS-FPN model with several classical and recently developed deep learning models, using the high-resolution SAR images dataset (HRSID) that contains multi-scale marine ship SAR images, demonstrated the superior performance of MS-FPN over other models.https://www.mdpi.com/2072-4292/14/24/6312multi-scale feature pyramid network (MS-FPN)multi-scale feature mapsship detectionship segmentationsynthetic aperture radar (SAR) |
| spellingShingle | Zequn Sun Chunning Meng Jierong Cheng Zhiqing Zhang Shengjiang Chang A Multi-Scale Feature Pyramid Network for Detection and Instance Segmentation of Marine Ships in SAR Images Remote Sensing multi-scale feature pyramid network (MS-FPN) multi-scale feature maps ship detection ship segmentation synthetic aperture radar (SAR) |
| title | A Multi-Scale Feature Pyramid Network for Detection and Instance Segmentation of Marine Ships in SAR Images |
| title_full | A Multi-Scale Feature Pyramid Network for Detection and Instance Segmentation of Marine Ships in SAR Images |
| title_fullStr | A Multi-Scale Feature Pyramid Network for Detection and Instance Segmentation of Marine Ships in SAR Images |
| title_full_unstemmed | A Multi-Scale Feature Pyramid Network for Detection and Instance Segmentation of Marine Ships in SAR Images |
| title_short | A Multi-Scale Feature Pyramid Network for Detection and Instance Segmentation of Marine Ships in SAR Images |
| title_sort | multi scale feature pyramid network for detection and instance segmentation of marine ships in sar images |
| topic | multi-scale feature pyramid network (MS-FPN) multi-scale feature maps ship detection ship segmentation synthetic aperture radar (SAR) |
| url | https://www.mdpi.com/2072-4292/14/24/6312 |
| work_keys_str_mv | AT zequnsun amultiscalefeaturepyramidnetworkfordetectionandinstancesegmentationofmarineshipsinsarimages AT chunningmeng amultiscalefeaturepyramidnetworkfordetectionandinstancesegmentationofmarineshipsinsarimages AT jierongcheng amultiscalefeaturepyramidnetworkfordetectionandinstancesegmentationofmarineshipsinsarimages AT zhiqingzhang amultiscalefeaturepyramidnetworkfordetectionandinstancesegmentationofmarineshipsinsarimages AT shengjiangchang amultiscalefeaturepyramidnetworkfordetectionandinstancesegmentationofmarineshipsinsarimages AT zequnsun multiscalefeaturepyramidnetworkfordetectionandinstancesegmentationofmarineshipsinsarimages AT chunningmeng multiscalefeaturepyramidnetworkfordetectionandinstancesegmentationofmarineshipsinsarimages AT jierongcheng multiscalefeaturepyramidnetworkfordetectionandinstancesegmentationofmarineshipsinsarimages AT zhiqingzhang multiscalefeaturepyramidnetworkfordetectionandinstancesegmentationofmarineshipsinsarimages AT shengjiangchang multiscalefeaturepyramidnetworkfordetectionandinstancesegmentationofmarineshipsinsarimages |