Shape-Constrained Method of Remote Sensing Monitoring of Marine Raft Aquaculture Areas on Multitemporal Synthetic Sentinel-1 Imagery
Large-scale and periodic remote sensing monitoring of marine raft aquaculture areas is significant for scientific planning of their layout and for promoting sustainable development of marine ecology. Synthetic aperture radar (SAR) is an important tool for stable monitoring of marine raft aquaculture...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-03-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/5/1249 |
| _version_ | 1797473848147312640 |
|---|---|
| author | Yi Zhang Chengyi Wang Jingbo Chen Futao Wang |
| author_facet | Yi Zhang Chengyi Wang Jingbo Chen Futao Wang |
| author_sort | Yi Zhang |
| collection | DOAJ |
| description | Large-scale and periodic remote sensing monitoring of marine raft aquaculture areas is significant for scientific planning of their layout and for promoting sustainable development of marine ecology. Synthetic aperture radar (SAR) is an important tool for stable monitoring of marine raft aquaculture areas since it is all-weather, all-day, and cloud-penetrating. However, the scattering signal of marine raft aquaculture areas is affected by speckle noise and sea state, so their features in SAR images are complex. Thus, it is challenging to extract marine raft aquaculture areas from SAR images. In this paper, we propose a method to extract marine raft aquaculture areas from Sentinel-1 images based on the analysis of the features for marine raft aquaculture areas. First, the data are preprocessed using multitemporal phase synthesis to weaken the noise interference, enhance the signal of marine raft aquaculture areas, and improve the significance of the characteristics of raft aquaculture areas. Second, the geometric features of the marine raft aquaculture area are combined to design the model structure and introduce the shape constraint module, which adds a priori knowledge to guide the model convergence direction during the training process. Experiments verify that the method outperforms the popular semantic segmentation model with an <i>F</i>1 of 84.52%. |
| first_indexed | 2024-03-09T20:22:23Z |
| format | Article |
| id | doaj.art-795b9922fd914fd4946d942115afa880 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T20:22:23Z |
| publishDate | 2022-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-795b9922fd914fd4946d942115afa8802023-11-23T23:43:47ZengMDPI AGRemote Sensing2072-42922022-03-01145124910.3390/rs14051249Shape-Constrained Method of Remote Sensing Monitoring of Marine Raft Aquaculture Areas on Multitemporal Synthetic Sentinel-1 ImageryYi Zhang0Chengyi Wang1Jingbo Chen2Futao Wang3Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaLarge-scale and periodic remote sensing monitoring of marine raft aquaculture areas is significant for scientific planning of their layout and for promoting sustainable development of marine ecology. Synthetic aperture radar (SAR) is an important tool for stable monitoring of marine raft aquaculture areas since it is all-weather, all-day, and cloud-penetrating. However, the scattering signal of marine raft aquaculture areas is affected by speckle noise and sea state, so their features in SAR images are complex. Thus, it is challenging to extract marine raft aquaculture areas from SAR images. In this paper, we propose a method to extract marine raft aquaculture areas from Sentinel-1 images based on the analysis of the features for marine raft aquaculture areas. First, the data are preprocessed using multitemporal phase synthesis to weaken the noise interference, enhance the signal of marine raft aquaculture areas, and improve the significance of the characteristics of raft aquaculture areas. Second, the geometric features of the marine raft aquaculture area are combined to design the model structure and introduce the shape constraint module, which adds a priori knowledge to guide the model convergence direction during the training process. Experiments verify that the method outperforms the popular semantic segmentation model with an <i>F</i>1 of 84.52%.https://www.mdpi.com/2072-4292/14/5/1249monitoring of maricultureSARimage synthesissemantic segmentation |
| spellingShingle | Yi Zhang Chengyi Wang Jingbo Chen Futao Wang Shape-Constrained Method of Remote Sensing Monitoring of Marine Raft Aquaculture Areas on Multitemporal Synthetic Sentinel-1 Imagery Remote Sensing monitoring of mariculture SAR image synthesis semantic segmentation |
| title | Shape-Constrained Method of Remote Sensing Monitoring of Marine Raft Aquaculture Areas on Multitemporal Synthetic Sentinel-1 Imagery |
| title_full | Shape-Constrained Method of Remote Sensing Monitoring of Marine Raft Aquaculture Areas on Multitemporal Synthetic Sentinel-1 Imagery |
| title_fullStr | Shape-Constrained Method of Remote Sensing Monitoring of Marine Raft Aquaculture Areas on Multitemporal Synthetic Sentinel-1 Imagery |
| title_full_unstemmed | Shape-Constrained Method of Remote Sensing Monitoring of Marine Raft Aquaculture Areas on Multitemporal Synthetic Sentinel-1 Imagery |
| title_short | Shape-Constrained Method of Remote Sensing Monitoring of Marine Raft Aquaculture Areas on Multitemporal Synthetic Sentinel-1 Imagery |
| title_sort | shape constrained method of remote sensing monitoring of marine raft aquaculture areas on multitemporal synthetic sentinel 1 imagery |
| topic | monitoring of mariculture SAR image synthesis semantic segmentation |
| url | https://www.mdpi.com/2072-4292/14/5/1249 |
| work_keys_str_mv | AT yizhang shapeconstrainedmethodofremotesensingmonitoringofmarineraftaquacultureareasonmultitemporalsyntheticsentinel1imagery AT chengyiwang shapeconstrainedmethodofremotesensingmonitoringofmarineraftaquacultureareasonmultitemporalsyntheticsentinel1imagery AT jingbochen shapeconstrainedmethodofremotesensingmonitoringofmarineraftaquacultureareasonmultitemporalsyntheticsentinel1imagery AT futaowang shapeconstrainedmethodofremotesensingmonitoringofmarineraftaquacultureareasonmultitemporalsyntheticsentinel1imagery |