Multi-dimensional variables and feature parameter selection for aboveground biomass estimation of potato based on UAV multispectral imagery
Aboveground biomass (AGB) is an essential assessment of plant development and guiding agricultural production management in the field. Therefore, efficient and accurate access to crop AGB information can provide a timely and precise yield estimation, which is strong evidence for securing food supply...
Main Authors: | , , , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2022-07-01
|
Series: | Frontiers in Plant Science |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2022.948249/full |
_version_ | 1811223109372477440 |
---|---|
author | Shanjun Luo Shanjun Luo Shanjun Luo Xueqin Jiang Yingbin He Yingbin He Jianping Li Weihua Jiao Shengli Zhang Fei Xu Zhongcai Han Jing Sun Jinpeng Yang Xiangyi Wang Xintian Ma Zeru Lin |
author_facet | Shanjun Luo Shanjun Luo Shanjun Luo Xueqin Jiang Yingbin He Yingbin He Jianping Li Weihua Jiao Shengli Zhang Fei Xu Zhongcai Han Jing Sun Jinpeng Yang Xiangyi Wang Xintian Ma Zeru Lin |
author_sort | Shanjun Luo |
collection | DOAJ |
description | Aboveground biomass (AGB) is an essential assessment of plant development and guiding agricultural production management in the field. Therefore, efficient and accurate access to crop AGB information can provide a timely and precise yield estimation, which is strong evidence for securing food supply and trade. In this study, the spectral, texture, geometric, and frequency-domain variables were extracted through multispectral imagery of drones, and each variable importance for different dimensional parameter combinations was computed by three feature parameter selection methods. The selected variables from the different combinations were used to perform potato AGB estimation. The results showed that compared with no feature parameter selection, the accuracy and robustness of the AGB prediction models were significantly improved after parameter selection. The random forest based on out-of-bag (RF-OOB) method was proved to be the most effective feature selection method, and in combination with RF regression, the coefficient of determination (R2) of the AGB validation model could reach 0.90, with root mean square error (RMSE), mean absolute error (MAE), and normalized RMSE (nRMSE) of 71.68 g/m2, 51.27 g/m2, and 11.56%, respectively. Meanwhile, the regression models of the RF-OOB method provided a good solution to the problem that high AGB values were underestimated with the variables of four dimensions. Moreover, the precision of AGB estimates was improved as the dimensionality of parameters increased. This present work can contribute to a rapid, efficient, and non-destructive means of obtaining AGB information for crops as well as provide technical support for high-throughput plant phenotypes screening. |
first_indexed | 2024-04-12T08:26:28Z |
format | Article |
id | doaj.art-5fb6c783729443ef8e0489d7da3e57e9 |
institution | Directory Open Access Journal |
issn | 1664-462X |
language | English |
last_indexed | 2024-04-12T08:26:28Z |
publishDate | 2022-07-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Plant Science |
spelling | doaj.art-5fb6c783729443ef8e0489d7da3e57e92022-12-22T03:40:21ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2022-07-011310.3389/fpls.2022.948249948249Multi-dimensional variables and feature parameter selection for aboveground biomass estimation of potato based on UAV multispectral imageryShanjun Luo0Shanjun Luo1Shanjun Luo2Xueqin Jiang3Yingbin He4Yingbin He5Jianping Li6Weihua Jiao7Shengli Zhang8Fei Xu9Zhongcai Han10Jing Sun11Jinpeng Yang12Xiangyi Wang13Xintian Ma14Zeru Lin15Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing, ChinaCollaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science & Technology, Nanjing, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaInstitute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing, ChinaCollaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science & Technology, Nanjing, ChinaInstitute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing, ChinaCenter for Agricultural and Rural Economic Research, Shandong University of Finance and Economics, Jinan, ChinaPotato Science Institute, Jilin Academy of Vegetables and Flower Sciences, Changchun, ChinaPotato Science Institute, Jilin Academy of Vegetables and Flower Sciences, Changchun, ChinaPotato Science Institute, Jilin Academy of Vegetables and Flower Sciences, Changchun, ChinaPotato Science Institute, Jilin Academy of Vegetables and Flower Sciences, Changchun, ChinaInstitute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing, ChinaInstitute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing, ChinaInstitute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing, ChinaSchool of Economics and Management, Tiangong University, Tianjin, ChinaAboveground biomass (AGB) is an essential assessment of plant development and guiding agricultural production management in the field. Therefore, efficient and accurate access to crop AGB information can provide a timely and precise yield estimation, which is strong evidence for securing food supply and trade. In this study, the spectral, texture, geometric, and frequency-domain variables were extracted through multispectral imagery of drones, and each variable importance for different dimensional parameter combinations was computed by three feature parameter selection methods. The selected variables from the different combinations were used to perform potato AGB estimation. The results showed that compared with no feature parameter selection, the accuracy and robustness of the AGB prediction models were significantly improved after parameter selection. The random forest based on out-of-bag (RF-OOB) method was proved to be the most effective feature selection method, and in combination with RF regression, the coefficient of determination (R2) of the AGB validation model could reach 0.90, with root mean square error (RMSE), mean absolute error (MAE), and normalized RMSE (nRMSE) of 71.68 g/m2, 51.27 g/m2, and 11.56%, respectively. Meanwhile, the regression models of the RF-OOB method provided a good solution to the problem that high AGB values were underestimated with the variables of four dimensions. Moreover, the precision of AGB estimates was improved as the dimensionality of parameters increased. This present work can contribute to a rapid, efficient, and non-destructive means of obtaining AGB information for crops as well as provide technical support for high-throughput plant phenotypes screening.https://www.frontiersin.org/articles/10.3389/fpls.2022.948249/fullremote sensing phenotypesspectral indicestexturegeometric parametersfrequency-domain indicatorsvariables preference |
spellingShingle | Shanjun Luo Shanjun Luo Shanjun Luo Xueqin Jiang Yingbin He Yingbin He Jianping Li Weihua Jiao Shengli Zhang Fei Xu Zhongcai Han Jing Sun Jinpeng Yang Xiangyi Wang Xintian Ma Zeru Lin Multi-dimensional variables and feature parameter selection for aboveground biomass estimation of potato based on UAV multispectral imagery Frontiers in Plant Science remote sensing phenotypes spectral indices texture geometric parameters frequency-domain indicators variables preference |
title | Multi-dimensional variables and feature parameter selection for aboveground biomass estimation of potato based on UAV multispectral imagery |
title_full | Multi-dimensional variables and feature parameter selection for aboveground biomass estimation of potato based on UAV multispectral imagery |
title_fullStr | Multi-dimensional variables and feature parameter selection for aboveground biomass estimation of potato based on UAV multispectral imagery |
title_full_unstemmed | Multi-dimensional variables and feature parameter selection for aboveground biomass estimation of potato based on UAV multispectral imagery |
title_short | Multi-dimensional variables and feature parameter selection for aboveground biomass estimation of potato based on UAV multispectral imagery |
title_sort | multi dimensional variables and feature parameter selection for aboveground biomass estimation of potato based on uav multispectral imagery |
topic | remote sensing phenotypes spectral indices texture geometric parameters frequency-domain indicators variables preference |
url | https://www.frontiersin.org/articles/10.3389/fpls.2022.948249/full |
work_keys_str_mv | AT shanjunluo multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT shanjunluo multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT shanjunluo multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT xueqinjiang multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT yingbinhe multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT yingbinhe multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT jianpingli multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT weihuajiao multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT shenglizhang multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT feixu multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT zhongcaihan multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT jingsun multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT jinpengyang multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT xiangyiwang multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT xintianma multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery AT zerulin multidimensionalvariablesandfeatureparameterselectionforabovegroundbiomassestimationofpotatobasedonuavmultispectralimagery |