Semantic segmentation for plastic-covered greenhouses and plastic-mulched farmlands from VHR imagery
ABSTRACTDue to their important role in maintaining temperature and soil moisture, agricultural plastic covers have been widely utilized around the globe for improving crop-growing conditions, which include both plastic-covered greenhouses (PCGs) and plastic-mulched farmlands (PMFs). However, it is a...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2023-12-01
|
| Series: | International Journal of Digital Earth |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/17538947.2023.2275657 |
| _version_ | 1797635764333314048 |
|---|---|
| author | Bowen Niu Quanlong Feng Shuai Su Zhi Yang Sihang Zhang Shaotong Liu Jiudong Wang Jianyu Yang Jianhua Gong |
| author_facet | Bowen Niu Quanlong Feng Shuai Su Zhi Yang Sihang Zhang Shaotong Liu Jiudong Wang Jianyu Yang Jianhua Gong |
| author_sort | Bowen Niu |
| collection | DOAJ |
| description | ABSTRACTDue to their important role in maintaining temperature and soil moisture, agricultural plastic covers have been widely utilized around the globe for improving crop-growing conditions, which include both plastic-covered greenhouses (PCGs) and plastic-mulched farmlands (PMFs). However, it is a challenging and long-neglected issue to separate PCGs from PMFs due to their spectral similarity. The objective of this study is to propose a deep semantic segmentation model for accurate PCG and PMF mapping based on very high-resolution satellite images and to improve the model’s spatial generalization capability using a transfer learning strategy. Specifically, the proposed semantic segmentation model has an encoder-decoder structure, where the encoder is composed of a new convolutional neural network for discriminative spatial feature learning, while the decoder utilizes a multi-task strategy to improve the predictions on the boundaries. Meanwhile, a transfer learning framework is adopted to increase mapping performance and generalization ability under limited samples. Experimental results in several typical regions across the Eurasian continent show that the proposed model could separate PCGs from PMFs accurately with a mean overall accuracy of 94.49% and an average mIoU of 0.8377. Ablation studies verify the role of encoder-decoder and transfer learning strategy in improving classification performance. |
| first_indexed | 2024-03-11T12:25:58Z |
| format | Article |
| id | doaj.art-afe2d49448124963b013141a5cabafd0 |
| institution | Directory Open Access Journal |
| issn | 1753-8947 1753-8955 |
| language | English |
| last_indexed | 2024-03-11T12:25:58Z |
| publishDate | 2023-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | International Journal of Digital Earth |
| spelling | doaj.art-afe2d49448124963b013141a5cabafd02023-11-06T09:41:57ZengTaylor & Francis GroupInternational Journal of Digital Earth1753-89471753-89552023-12-011624553457210.1080/17538947.2023.2275657Semantic segmentation for plastic-covered greenhouses and plastic-mulched farmlands from VHR imageryBowen Niu0Quanlong Feng1Shuai Su2Zhi Yang3Sihang Zhang4Shaotong Liu5Jiudong Wang6Jianyu Yang7Jianhua Gong8College of Land Science and Technology, China Agricultural University, Beijing, People’s Republic of ChinaCollege of Land Science and Technology, China Agricultural University, Beijing, People’s Republic of ChinaCollege of Land Science and Technology, China Agricultural University, Beijing, People’s Republic of ChinaChina Electric Power Research Institute, Beijing, People’s Republic of ChinaChina Electric Power Research Institute, Beijing, People’s Republic of ChinaCollege of Land Science and Technology, China Agricultural University, Beijing, People’s Republic of ChinaCollege of Land Science and Technology, China Agricultural University, Beijing, People’s Republic of ChinaCollege of Land Science and Technology, China Agricultural University, Beijing, People’s Republic of ChinaNational Engineering Research Center for Geoinformatics, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, People’s Republic of ChinaABSTRACTDue to their important role in maintaining temperature and soil moisture, agricultural plastic covers have been widely utilized around the globe for improving crop-growing conditions, which include both plastic-covered greenhouses (PCGs) and plastic-mulched farmlands (PMFs). However, it is a challenging and long-neglected issue to separate PCGs from PMFs due to their spectral similarity. The objective of this study is to propose a deep semantic segmentation model for accurate PCG and PMF mapping based on very high-resolution satellite images and to improve the model’s spatial generalization capability using a transfer learning strategy. Specifically, the proposed semantic segmentation model has an encoder-decoder structure, where the encoder is composed of a new convolutional neural network for discriminative spatial feature learning, while the decoder utilizes a multi-task strategy to improve the predictions on the boundaries. Meanwhile, a transfer learning framework is adopted to increase mapping performance and generalization ability under limited samples. Experimental results in several typical regions across the Eurasian continent show that the proposed model could separate PCGs from PMFs accurately with a mean overall accuracy of 94.49% and an average mIoU of 0.8377. Ablation studies verify the role of encoder-decoder and transfer learning strategy in improving classification performance.https://www.tandfonline.com/doi/10.1080/17538947.2023.2275657Agricultural plastic coversdeep learningsemantic segmentationsatellite remote sensingtransfer learning |
| spellingShingle | Bowen Niu Quanlong Feng Shuai Su Zhi Yang Sihang Zhang Shaotong Liu Jiudong Wang Jianyu Yang Jianhua Gong Semantic segmentation for plastic-covered greenhouses and plastic-mulched farmlands from VHR imagery International Journal of Digital Earth Agricultural plastic covers deep learning semantic segmentation satellite remote sensing transfer learning |
| title | Semantic segmentation for plastic-covered greenhouses and plastic-mulched farmlands from VHR imagery |
| title_full | Semantic segmentation for plastic-covered greenhouses and plastic-mulched farmlands from VHR imagery |
| title_fullStr | Semantic segmentation for plastic-covered greenhouses and plastic-mulched farmlands from VHR imagery |
| title_full_unstemmed | Semantic segmentation for plastic-covered greenhouses and plastic-mulched farmlands from VHR imagery |
| title_short | Semantic segmentation for plastic-covered greenhouses and plastic-mulched farmlands from VHR imagery |
| title_sort | semantic segmentation for plastic covered greenhouses and plastic mulched farmlands from vhr imagery |
| topic | Agricultural plastic covers deep learning semantic segmentation satellite remote sensing transfer learning |
| url | https://www.tandfonline.com/doi/10.1080/17538947.2023.2275657 |
| work_keys_str_mv | AT bowenniu semanticsegmentationforplasticcoveredgreenhousesandplasticmulchedfarmlandsfromvhrimagery AT quanlongfeng semanticsegmentationforplasticcoveredgreenhousesandplasticmulchedfarmlandsfromvhrimagery AT shuaisu semanticsegmentationforplasticcoveredgreenhousesandplasticmulchedfarmlandsfromvhrimagery AT zhiyang semanticsegmentationforplasticcoveredgreenhousesandplasticmulchedfarmlandsfromvhrimagery AT sihangzhang semanticsegmentationforplasticcoveredgreenhousesandplasticmulchedfarmlandsfromvhrimagery AT shaotongliu semanticsegmentationforplasticcoveredgreenhousesandplasticmulchedfarmlandsfromvhrimagery AT jiudongwang semanticsegmentationforplasticcoveredgreenhousesandplasticmulchedfarmlandsfromvhrimagery AT jianyuyang semanticsegmentationforplasticcoveredgreenhousesandplasticmulchedfarmlandsfromvhrimagery AT jianhuagong semanticsegmentationforplasticcoveredgreenhousesandplasticmulchedfarmlandsfromvhrimagery |