Deriving Agricultural Field Boundaries for Crop Management from Satellite Images Using Semantic Feature Pyramid Network
We propose a Semantic Feature Pyramid Network (FPN)-based algorithm to derive agricultural field boundaries and internal non-planting regions from satellite imagery. It is aimed at providing guidance not only for land use management, but more importantly for harvest or crop protection machinery plan...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-06-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/11/2937 |
| _version_ | 1827739242829185024 |
|---|---|
| author | Yang Xu Xinyu Xue Zhu Sun Wei Gu Longfei Cui Yongkui Jin Yubin Lan |
| author_facet | Yang Xu Xinyu Xue Zhu Sun Wei Gu Longfei Cui Yongkui Jin Yubin Lan |
| author_sort | Yang Xu |
| collection | DOAJ |
| description | We propose a Semantic Feature Pyramid Network (FPN)-based algorithm to derive agricultural field boundaries and internal non-planting regions from satellite imagery. It is aimed at providing guidance not only for land use management, but more importantly for harvest or crop protection machinery planning. The Semantic Convolutional Neural Network (CNN) FPN is first employed for pixel-wise classification on each remote sensing image, detecting agricultural parcels; a post-processing method is then developed to transfer attained pixel classification results into closed contours, as field boundaries and internal non-planting regions, including slender paths (walking or water) and obstacles (trees or electronic poles). Three study sites with different plot sizes (0.11 ha, 1.39 ha, and 2.24 ha) are selected to validate the effectiveness of our algorithm, and the performance compared with other semantic CNN (including U-Net, U-Net++, PSP-Net, and Link-Net)-based algorithms. The test results show that the crop acreage information, field boundaries, and internal non-planting area could be determined by using the proposed algorithm in different places. When the boundary number applicable for machinery planning is attained, average and total crop planting area values all remain closer to the reference ones generally when using the semantic FPN with post-processing, compared with other methods. The post-processing methodology would greatly decrease the number of inapplicable and redundant field boundaries for path planning using different CNN models. In addition, the crop planting mode and scale (especially the small-scale planting and small/blurred gap between fields) both make a great difference to the boundary delineation and crop acreage determination. |
| first_indexed | 2024-03-11T02:58:03Z |
| format | Article |
| id | doaj.art-cdec17640c1044f89c6a6aac96e9b196 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T02:58:03Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-cdec17640c1044f89c6a6aac96e9b1962023-11-18T08:30:46ZengMDPI AGRemote Sensing2072-42922023-06-011511293710.3390/rs15112937Deriving Agricultural Field Boundaries for Crop Management from Satellite Images Using Semantic Feature Pyramid NetworkYang Xu0Xinyu Xue1Zhu Sun2Wei Gu3Longfei Cui4Yongkui Jin5Yubin Lan6Nanjing Institute of Agriculture Mechanization, Ministry of Agriculture and Rural Affairs, Nanjing 210014, ChinaNanjing Institute of Agriculture Mechanization, Ministry of Agriculture and Rural Affairs, Nanjing 210014, ChinaNanjing Institute of Agriculture Mechanization, Ministry of Agriculture and Rural Affairs, Nanjing 210014, ChinaNanjing Institute of Agriculture Mechanization, Ministry of Agriculture and Rural Affairs, Nanjing 210014, ChinaNanjing Institute of Agriculture Mechanization, Ministry of Agriculture and Rural Affairs, Nanjing 210014, ChinaNanjing Institute of Agriculture Mechanization, Ministry of Agriculture and Rural Affairs, Nanjing 210014, ChinaCollege of Electronic Engineering, South China Agricultural University, Guangzhou 510642, ChinaWe propose a Semantic Feature Pyramid Network (FPN)-based algorithm to derive agricultural field boundaries and internal non-planting regions from satellite imagery. It is aimed at providing guidance not only for land use management, but more importantly for harvest or crop protection machinery planning. The Semantic Convolutional Neural Network (CNN) FPN is first employed for pixel-wise classification on each remote sensing image, detecting agricultural parcels; a post-processing method is then developed to transfer attained pixel classification results into closed contours, as field boundaries and internal non-planting regions, including slender paths (walking or water) and obstacles (trees or electronic poles). Three study sites with different plot sizes (0.11 ha, 1.39 ha, and 2.24 ha) are selected to validate the effectiveness of our algorithm, and the performance compared with other semantic CNN (including U-Net, U-Net++, PSP-Net, and Link-Net)-based algorithms. The test results show that the crop acreage information, field boundaries, and internal non-planting area could be determined by using the proposed algorithm in different places. When the boundary number applicable for machinery planning is attained, average and total crop planting area values all remain closer to the reference ones generally when using the semantic FPN with post-processing, compared with other methods. The post-processing methodology would greatly decrease the number of inapplicable and redundant field boundaries for path planning using different CNN models. In addition, the crop planting mode and scale (especially the small-scale planting and small/blurred gap between fields) both make a great difference to the boundary delineation and crop acreage determination.https://www.mdpi.com/2072-4292/15/11/2937field boundary delineationinternal non-planting region detectionsatellite imageryland use guidanceagricultural machinery planningcrop acreage |
| spellingShingle | Yang Xu Xinyu Xue Zhu Sun Wei Gu Longfei Cui Yongkui Jin Yubin Lan Deriving Agricultural Field Boundaries for Crop Management from Satellite Images Using Semantic Feature Pyramid Network Remote Sensing field boundary delineation internal non-planting region detection satellite imagery land use guidance agricultural machinery planning crop acreage |
| title | Deriving Agricultural Field Boundaries for Crop Management from Satellite Images Using Semantic Feature Pyramid Network |
| title_full | Deriving Agricultural Field Boundaries for Crop Management from Satellite Images Using Semantic Feature Pyramid Network |
| title_fullStr | Deriving Agricultural Field Boundaries for Crop Management from Satellite Images Using Semantic Feature Pyramid Network |
| title_full_unstemmed | Deriving Agricultural Field Boundaries for Crop Management from Satellite Images Using Semantic Feature Pyramid Network |
| title_short | Deriving Agricultural Field Boundaries for Crop Management from Satellite Images Using Semantic Feature Pyramid Network |
| title_sort | deriving agricultural field boundaries for crop management from satellite images using semantic feature pyramid network |
| topic | field boundary delineation internal non-planting region detection satellite imagery land use guidance agricultural machinery planning crop acreage |
| url | https://www.mdpi.com/2072-4292/15/11/2937 |
| work_keys_str_mv | AT yangxu derivingagriculturalfieldboundariesforcropmanagementfromsatelliteimagesusingsemanticfeaturepyramidnetwork AT xinyuxue derivingagriculturalfieldboundariesforcropmanagementfromsatelliteimagesusingsemanticfeaturepyramidnetwork AT zhusun derivingagriculturalfieldboundariesforcropmanagementfromsatelliteimagesusingsemanticfeaturepyramidnetwork AT weigu derivingagriculturalfieldboundariesforcropmanagementfromsatelliteimagesusingsemanticfeaturepyramidnetwork AT longfeicui derivingagriculturalfieldboundariesforcropmanagementfromsatelliteimagesusingsemanticfeaturepyramidnetwork AT yongkuijin derivingagriculturalfieldboundariesforcropmanagementfromsatelliteimagesusingsemanticfeaturepyramidnetwork AT yubinlan derivingagriculturalfieldboundariesforcropmanagementfromsatelliteimagesusingsemanticfeaturepyramidnetwork |