Decomposition-Based Soil Moisture Estimation Using UAVSAR Fully Polarimetric Images
Polarimetric decomposition extracts scattering features that are indicative of the physical characteristics of the target. In this study, three polarimetric decomposition methods were tested for soil moisture estimation over agricultural fields using machine learning algorithms. Features extracted f...
| Asıl Yazarlar: | , , , |
|---|---|
| Materyal Türü: | Makale |
| Dil: | English |
| Baskı/Yayın Bilgisi: |
MDPI AG
2021-01-01
|
| Seri Bilgileri: | Agronomy |
| Konular: | |
| Online Erişim: | https://www.mdpi.com/2073-4395/11/1/145 |
| _version_ | 1827601174174367744 |
|---|---|
| author | Zeinab Akhavan Mahdi Hasanlou Mehdi Hosseini Heather McNairn |
| author_facet | Zeinab Akhavan Mahdi Hasanlou Mehdi Hosseini Heather McNairn |
| author_sort | Zeinab Akhavan |
| collection | DOAJ |
| description | Polarimetric decomposition extracts scattering features that are indicative of the physical characteristics of the target. In this study, three polarimetric decomposition methods were tested for soil moisture estimation over agricultural fields using machine learning algorithms. Features extracted from model-based Freeman–Durden, Eigenvalue and Eigenvector based H/A/α, and Van Zyl decompositions were used as inputs in random forest and neural network regression algorithms. These algorithms were applied to retrieve soil moisture over soybean, wheat, and corn fields. A time series of polarimetric Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) data acquired during the Soil Moisture Active Passive Experiment 2012 (SMAPVEX12) field campaign was used for the training and validation of the algorithms. Three feature selection methods were tested to determine the best input features for the machine learning algorithms. The most accurate soil moisture estimates were derived from the random forest regression algorithm for soybeans, with a correlation of determination (R<sup>2</sup>) of 0.86, root mean square error (RMSE) of 0.041 m<sup>3</sup> m<sup>−3</sup> and mean absolute error (MAE) of 0.030 m<sup>3</sup> m<sup>−3</sup>. Feature selection also impacted results. Some features like anisotropy, Horizontal transmit and Horizontal receive (HH), and surface roughness parameters (correlation length and RMS-H) had a direct effect on all algorithm performance enhancement as these parameters have a direct impact on the backscattered signal. |
| first_indexed | 2024-03-09T04:50:29Z |
| format | Article |
| id | doaj.art-ff98a5bea41a4c2a89a14e547d2d01cf |
| institution | Directory Open Access Journal |
| issn | 2073-4395 |
| language | English |
| last_indexed | 2024-03-09T04:50:29Z |
| publishDate | 2021-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agronomy |
| spelling | doaj.art-ff98a5bea41a4c2a89a14e547d2d01cf2023-12-03T13:11:30ZengMDPI AGAgronomy2073-43952021-01-0111114510.3390/agronomy11010145Decomposition-Based Soil Moisture Estimation Using UAVSAR Fully Polarimetric ImagesZeinab Akhavan0Mahdi Hasanlou1Mehdi Hosseini2Heather McNairn3School of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Tehran 1417466191, IranSchool of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Tehran 1417466191, IranDepartment of Geographical Sciences, University of Maryland, College Park, MD 20742, USAAgriculture and Agri-Food Canada, 960 Carling Ave., Ottawa, ON K1A 0C6, CanadaPolarimetric decomposition extracts scattering features that are indicative of the physical characteristics of the target. In this study, three polarimetric decomposition methods were tested for soil moisture estimation over agricultural fields using machine learning algorithms. Features extracted from model-based Freeman–Durden, Eigenvalue and Eigenvector based H/A/α, and Van Zyl decompositions were used as inputs in random forest and neural network regression algorithms. These algorithms were applied to retrieve soil moisture over soybean, wheat, and corn fields. A time series of polarimetric Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) data acquired during the Soil Moisture Active Passive Experiment 2012 (SMAPVEX12) field campaign was used for the training and validation of the algorithms. Three feature selection methods were tested to determine the best input features for the machine learning algorithms. The most accurate soil moisture estimates were derived from the random forest regression algorithm for soybeans, with a correlation of determination (R<sup>2</sup>) of 0.86, root mean square error (RMSE) of 0.041 m<sup>3</sup> m<sup>−3</sup> and mean absolute error (MAE) of 0.030 m<sup>3</sup> m<sup>−3</sup>. Feature selection also impacted results. Some features like anisotropy, Horizontal transmit and Horizontal receive (HH), and surface roughness parameters (correlation length and RMS-H) had a direct effect on all algorithm performance enhancement as these parameters have a direct impact on the backscattered signal.https://www.mdpi.com/2073-4395/11/1/145soil moistureagriculturerandom forestneural networkSMAPVEX12UAVSAR |
| spellingShingle | Zeinab Akhavan Mahdi Hasanlou Mehdi Hosseini Heather McNairn Decomposition-Based Soil Moisture Estimation Using UAVSAR Fully Polarimetric Images Agronomy soil moisture agriculture random forest neural network SMAPVEX12 UAVSAR |
| title | Decomposition-Based Soil Moisture Estimation Using UAVSAR Fully Polarimetric Images |
| title_full | Decomposition-Based Soil Moisture Estimation Using UAVSAR Fully Polarimetric Images |
| title_fullStr | Decomposition-Based Soil Moisture Estimation Using UAVSAR Fully Polarimetric Images |
| title_full_unstemmed | Decomposition-Based Soil Moisture Estimation Using UAVSAR Fully Polarimetric Images |
| title_short | Decomposition-Based Soil Moisture Estimation Using UAVSAR Fully Polarimetric Images |
| title_sort | decomposition based soil moisture estimation using uavsar fully polarimetric images |
| topic | soil moisture agriculture random forest neural network SMAPVEX12 UAVSAR |
| url | https://www.mdpi.com/2073-4395/11/1/145 |
| work_keys_str_mv | AT zeinabakhavan decompositionbasedsoilmoistureestimationusinguavsarfullypolarimetricimages AT mahdihasanlou decompositionbasedsoilmoistureestimationusinguavsarfullypolarimetricimages AT mehdihosseini decompositionbasedsoilmoistureestimationusinguavsarfullypolarimetricimages AT heathermcnairn decompositionbasedsoilmoistureestimationusinguavsarfullypolarimetricimages |