The Application of an Unmanned Aerial System and Machine Learning Techniques for Red Clover-Grass Mixture Yield Estimation under Variety Performance Trials
A significant trend has developed with the recent growing interest in the estimation of aboveground biomass of vegetation in legume-supported systems in perennial or semi-natural grasslands to meet the demands of sustainable and precise agriculture. Unmanned aerial systems (UAS) are a powerful tool...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Sprog: | English |
| Udgivet: |
MDPI AG
2021-05-01
|
| Serier: | Remote Sensing |
| Fag: | |
| Online adgang: | https://www.mdpi.com/2072-4292/13/10/1994 |
| _version_ | 1827692080690888704 |
|---|---|
| author | Kai-Yun Li Niall G. Burnside Raul Sampaio de Lima Miguel Villoslada Peciña Karli Sepp Ming-Der Yang Janar Raet Ants Vain Are Selge Kalev Sepp |
| author_facet | Kai-Yun Li Niall G. Burnside Raul Sampaio de Lima Miguel Villoslada Peciña Karli Sepp Ming-Der Yang Janar Raet Ants Vain Are Selge Kalev Sepp |
| author_sort | Kai-Yun Li |
| collection | DOAJ |
| description | A significant trend has developed with the recent growing interest in the estimation of aboveground biomass of vegetation in legume-supported systems in perennial or semi-natural grasslands to meet the demands of sustainable and precise agriculture. Unmanned aerial systems (UAS) are a powerful tool when it comes to supporting farm-scale phenotyping trials. In this study, we explored the variation of the red clover-grass mixture dry matter (DM) yields between temporal periods (one- and two-year cultivated), farming operations [soil tillage methods (STM), cultivation methods (CM), manure application (MA)] using three machine learning (ML) techniques [random forest regression (RFR), support vector regression (SVR), and artificial neural network (ANN)] and six multispectral vegetation indices (VIs) to predict DM yields. The ML evaluation results showed the best performance for ANN in the 11-day before harvest category (R<sup>2</sup> = 0.90, NRMSE = 0.12), followed by RFR (R<sup>2</sup> = 0.90 NRMSE = 0.15), and SVR (R<sup>2</sup> = 0.86, NRMSE = 0.16), which was furthermore supported by the leave-one-out cross-validation pre-analysis. In terms of VI performance, green normalized difference vegetation index (GNDVI), green difference vegetation index (GDVI), as well as modified simple ratio (MSR) performed better as predictors in ANN and RFR. However, the prediction ability of models was being influenced by farming operations. The stratified sampling, based on STM, had a better model performance than CM and MA. It is proposed that drone data collection was suggested to be optimum in this study, closer to the harvest date, but not later than the ageing stage. |
| first_indexed | 2024-03-10T11:15:24Z |
| format | Article |
| id | doaj.art-a9af1140cac547bea6b9562e4e4d430a |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T11:15:24Z |
| publishDate | 2021-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-a9af1140cac547bea6b9562e4e4d430a2023-11-21T20:29:09ZengMDPI AGRemote Sensing2072-42922021-05-011310199410.3390/rs13101994The Application of an Unmanned Aerial System and Machine Learning Techniques for Red Clover-Grass Mixture Yield Estimation under Variety Performance TrialsKai-Yun Li0Niall G. Burnside1Raul Sampaio de Lima2Miguel Villoslada Peciña3Karli Sepp4Ming-Der Yang5Janar Raet6Ants Vain7Are Selge8Kalev Sepp9Institute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaSchool of Environment & Technology, University of Brighton, Lewes Road, Brighton BN2 4JG, UKInstitute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaInstitute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaAgricultural Research Center, 4/6 Teaduse St., 75501 Saku, EstoniaDepartment of Civil Engineering, and Innovation and Development Center of Sustainable Agriculture, National Chung Hsing University, Taichung 402, TaiwanInstitute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaInstitute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaInstitute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaInstitute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaA significant trend has developed with the recent growing interest in the estimation of aboveground biomass of vegetation in legume-supported systems in perennial or semi-natural grasslands to meet the demands of sustainable and precise agriculture. Unmanned aerial systems (UAS) are a powerful tool when it comes to supporting farm-scale phenotyping trials. In this study, we explored the variation of the red clover-grass mixture dry matter (DM) yields between temporal periods (one- and two-year cultivated), farming operations [soil tillage methods (STM), cultivation methods (CM), manure application (MA)] using three machine learning (ML) techniques [random forest regression (RFR), support vector regression (SVR), and artificial neural network (ANN)] and six multispectral vegetation indices (VIs) to predict DM yields. The ML evaluation results showed the best performance for ANN in the 11-day before harvest category (R<sup>2</sup> = 0.90, NRMSE = 0.12), followed by RFR (R<sup>2</sup> = 0.90 NRMSE = 0.15), and SVR (R<sup>2</sup> = 0.86, NRMSE = 0.16), which was furthermore supported by the leave-one-out cross-validation pre-analysis. In terms of VI performance, green normalized difference vegetation index (GNDVI), green difference vegetation index (GDVI), as well as modified simple ratio (MSR) performed better as predictors in ANN and RFR. However, the prediction ability of models was being influenced by farming operations. The stratified sampling, based on STM, had a better model performance than CM and MA. It is proposed that drone data collection was suggested to be optimum in this study, closer to the harvest date, but not later than the ageing stage.https://www.mdpi.com/2072-4292/13/10/1994unmanned aerial systemred cloverrandom forestsupport vector regressionartificial neural networktillage |
| spellingShingle | Kai-Yun Li Niall G. Burnside Raul Sampaio de Lima Miguel Villoslada Peciña Karli Sepp Ming-Der Yang Janar Raet Ants Vain Are Selge Kalev Sepp The Application of an Unmanned Aerial System and Machine Learning Techniques for Red Clover-Grass Mixture Yield Estimation under Variety Performance Trials Remote Sensing unmanned aerial system red clover random forest support vector regression artificial neural network tillage |
| title | The Application of an Unmanned Aerial System and Machine Learning Techniques for Red Clover-Grass Mixture Yield Estimation under Variety Performance Trials |
| title_full | The Application of an Unmanned Aerial System and Machine Learning Techniques for Red Clover-Grass Mixture Yield Estimation under Variety Performance Trials |
| title_fullStr | The Application of an Unmanned Aerial System and Machine Learning Techniques for Red Clover-Grass Mixture Yield Estimation under Variety Performance Trials |
| title_full_unstemmed | The Application of an Unmanned Aerial System and Machine Learning Techniques for Red Clover-Grass Mixture Yield Estimation under Variety Performance Trials |
| title_short | The Application of an Unmanned Aerial System and Machine Learning Techniques for Red Clover-Grass Mixture Yield Estimation under Variety Performance Trials |
| title_sort | application of an unmanned aerial system and machine learning techniques for red clover grass mixture yield estimation under variety performance trials |
| topic | unmanned aerial system red clover random forest support vector regression artificial neural network tillage |
| url | https://www.mdpi.com/2072-4292/13/10/1994 |
| work_keys_str_mv | AT kaiyunli theapplicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT niallgburnside theapplicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT raulsampaiodelima theapplicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT miguelvillosladapecina theapplicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT karlisepp theapplicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT mingderyang theapplicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT janarraet theapplicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT antsvain theapplicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT areselge theapplicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT kalevsepp theapplicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT kaiyunli applicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT niallgburnside applicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT raulsampaiodelima applicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT miguelvillosladapecina applicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT karlisepp applicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT mingderyang applicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT janarraet applicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT antsvain applicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT areselge applicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials AT kalevsepp applicationofanunmannedaerialsystemandmachinelearningtechniquesforredclovergrassmixtureyieldestimationundervarietyperformancetrials |