Using Machine Learning for Estimating Rice Chlorophyll Content from In Situ Hyperspectral Data
Chlorophyll is an essential pigment for photosynthesis in crops, and leaf chlorophyll content can be used as an indicator for crop growth status and help guide nitrogen fertilizer applications. Estimating crop chlorophyll content plays an important role in precision agriculture. In this study, a var...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/18/3104 |
| _version_ | 1827705704018870272 |
|---|---|
| author | Gangqiang An Minfeng Xing Binbin He Chunhua Liao Xiaodong Huang Jiali Shang Haiqi Kang |
| author_facet | Gangqiang An Minfeng Xing Binbin He Chunhua Liao Xiaodong Huang Jiali Shang Haiqi Kang |
| author_sort | Gangqiang An |
| collection | DOAJ |
| description | Chlorophyll is an essential pigment for photosynthesis in crops, and leaf chlorophyll content can be used as an indicator for crop growth status and help guide nitrogen fertilizer applications. Estimating crop chlorophyll content plays an important role in precision agriculture. In this study, a variable, rate of change in reflectance between wavelengths ‘a’ and ‘b’ (RCRW<sub>a-b</sub>), derived from in situ hyperspectral remote sensing data combined with four advanced machine learning techniques, Gaussian process regression (GPR), random forest regression (RFR), support vector regression (SVR), and gradient boosting regression tree (GBRT), were used to estimate the chlorophyll content (measured by a portable soil–plant analysis development meter) of rice. The performances of the four machine learning models were assessed and compared using root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R<sup>2</sup>). The results revealed that four features of RCRW<sub>a-b</sub>, RCRW<sub>551.0–565.6</sub>, RCRW<sub>739.5–743.5</sub>, RCRW<sub>684.4–687.1</sub> and RCRW<sub>667.9–672.0</sub>, were effective in estimating the chlorophyll content of rice, and the RFR model generated the highest prediction accuracy (training set: RMSE = 1.54, MAE =1.23 and R<sup>2</sup> = 0.95; validation set: RMSE = 2.64, MAE = 1.99 and R<sup>2</sup> = 0.80). The GPR model was found to have the strongest generalization (training set: RMSE = 2.83, MAE = 2.16 and R<sup>2</sup> = 0.77; validation set: RMSE = 2.97, MAE = 2.30 and R<sup>2</sup> = 0.76). We conclude that RCRW<sub>a-b</sub> is a useful variable to estimate chlorophyll content of rice, and RFR and GPR are powerful machine learning algorithms for estimating the chlorophyll content of rice. |
| first_indexed | 2024-03-10T16:08:37Z |
| format | Article |
| id | doaj.art-5582f00c1f42480ab9a1b79c9ae176be |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T16:08:37Z |
| publishDate | 2020-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-5582f00c1f42480ab9a1b79c9ae176be2023-11-20T14:39:43ZengMDPI AGRemote Sensing2072-42922020-09-011218310410.3390/rs12183104Using Machine Learning for Estimating Rice Chlorophyll Content from In Situ Hyperspectral DataGangqiang An0Minfeng Xing1Binbin He2Chunhua Liao3Xiaodong Huang4Jiali Shang5Haiqi Kang6School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, ChinaDepartment of Geography, Western University, London, ON N6A 5C2, CanadaApplied Geosolutions, 15 Newmarket Road, Durham, NH 03824, USAOttawa Research and Development Centre, Agriculture and Agri-Food Canada, 960 Carling Avenue, Ottawa, ON K1A 0C6, CanadaCrop Research Institute, Sichuan Academy of Agricultural Sciences, Chengdu 610066, ChinaChlorophyll is an essential pigment for photosynthesis in crops, and leaf chlorophyll content can be used as an indicator for crop growth status and help guide nitrogen fertilizer applications. Estimating crop chlorophyll content plays an important role in precision agriculture. In this study, a variable, rate of change in reflectance between wavelengths ‘a’ and ‘b’ (RCRW<sub>a-b</sub>), derived from in situ hyperspectral remote sensing data combined with four advanced machine learning techniques, Gaussian process regression (GPR), random forest regression (RFR), support vector regression (SVR), and gradient boosting regression tree (GBRT), were used to estimate the chlorophyll content (measured by a portable soil–plant analysis development meter) of rice. The performances of the four machine learning models were assessed and compared using root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R<sup>2</sup>). The results revealed that four features of RCRW<sub>a-b</sub>, RCRW<sub>551.0–565.6</sub>, RCRW<sub>739.5–743.5</sub>, RCRW<sub>684.4–687.1</sub> and RCRW<sub>667.9–672.0</sub>, were effective in estimating the chlorophyll content of rice, and the RFR model generated the highest prediction accuracy (training set: RMSE = 1.54, MAE =1.23 and R<sup>2</sup> = 0.95; validation set: RMSE = 2.64, MAE = 1.99 and R<sup>2</sup> = 0.80). The GPR model was found to have the strongest generalization (training set: RMSE = 2.83, MAE = 2.16 and R<sup>2</sup> = 0.77; validation set: RMSE = 2.97, MAE = 2.30 and R<sup>2</sup> = 0.76). We conclude that RCRW<sub>a-b</sub> is a useful variable to estimate chlorophyll content of rice, and RFR and GPR are powerful machine learning algorithms for estimating the chlorophyll content of rice.https://www.mdpi.com/2072-4292/12/18/3104hyperspectral remote sensingmachine learning technologyRCRW<sub>a-b</sub>SPAD valuerice |
| spellingShingle | Gangqiang An Minfeng Xing Binbin He Chunhua Liao Xiaodong Huang Jiali Shang Haiqi Kang Using Machine Learning for Estimating Rice Chlorophyll Content from In Situ Hyperspectral Data Remote Sensing hyperspectral remote sensing machine learning technology RCRW<sub>a-b</sub> SPAD value rice |
| title | Using Machine Learning for Estimating Rice Chlorophyll Content from In Situ Hyperspectral Data |
| title_full | Using Machine Learning for Estimating Rice Chlorophyll Content from In Situ Hyperspectral Data |
| title_fullStr | Using Machine Learning for Estimating Rice Chlorophyll Content from In Situ Hyperspectral Data |
| title_full_unstemmed | Using Machine Learning for Estimating Rice Chlorophyll Content from In Situ Hyperspectral Data |
| title_short | Using Machine Learning for Estimating Rice Chlorophyll Content from In Situ Hyperspectral Data |
| title_sort | using machine learning for estimating rice chlorophyll content from in situ hyperspectral data |
| topic | hyperspectral remote sensing machine learning technology RCRW<sub>a-b</sub> SPAD value rice |
| url | https://www.mdpi.com/2072-4292/12/18/3104 |
| work_keys_str_mv | AT gangqiangan usingmachinelearningforestimatingricechlorophyllcontentfrominsituhyperspectraldata AT minfengxing usingmachinelearningforestimatingricechlorophyllcontentfrominsituhyperspectraldata AT binbinhe usingmachinelearningforestimatingricechlorophyllcontentfrominsituhyperspectraldata AT chunhualiao usingmachinelearningforestimatingricechlorophyllcontentfrominsituhyperspectraldata AT xiaodonghuang usingmachinelearningforestimatingricechlorophyllcontentfrominsituhyperspectraldata AT jialishang usingmachinelearningforestimatingricechlorophyllcontentfrominsituhyperspectraldata AT haiqikang usingmachinelearningforestimatingricechlorophyllcontentfrominsituhyperspectraldata |