Entropy Weight Ensemble Framework for Yield Prediction of Winter Wheat Under Different Water Stress Treatments Using Unmanned Aerial Vehicle-Based Multispectral and Thermal Data

Abstract

Crop breeding programs generally perform early field assessments of candidate selection based on primary traits such as grain yield (GY). The traditional methods of yield assessment are costly, inefficient, and considered a bottleneck in modern precision agriculture. Recent advances in an unmanned aerial vehicle (UAV) and development of sensors have opened a new avenue for data acquisition cost-effectively and rapidly. We evaluated UAV-based multispectral and thermal images for in-season GY prediction using 30 winter wheat genotypes under 3 water treatments. For this, multispectral vegetation indices (VIs) and normalized relative canopy temperature (NRCT) were calculated and selected by the gray relational analysis (GRA) at each growth stage, i.e., jointing, booting, heading, flowering, grain filling, and maturity to reduce the data dimension. The elastic net regression (ENR) was developed by using selected features as input variables for yield prediction, whereas the entropy weight fusion (EWF) method was used to combine the predicted GY values from multiple growth stages. In our results, the fusion of dual-sensor data showed high yield prediction accuracy [coefficient of determination (R2) = 0.527–0.667] compared to using a single multispectral sensor (R2 = 0.130–0.461). Results showed that the grain filling stage was the optimal stage to predict GY with R2 = 0.667, root mean square error (RMSE) = 0.881 t ha–1, relative root-mean-square error (RRMSE) = 15.2%, and mean absolute error (MAE) = 0.721 t ha–1. The EWF model outperformed at all the individual growth stages with R2 varying from 0.677 to 0.729. The best prediction result (R2 = 0.729, RMSE = 0.831 t ha–1, RRMSE = 14.3%, and MAE = 0.684 t ha–1) was achieved through combining the predicted values of all growth stages. This study suggests that the fusion of UAV-based multispectral and thermal IR data within an ENR-EWF framework can provide a precise and robust prediction of wheat yield.

Rapid prediction of winter wheat yield and nitrogen use efficiency using consumer-grade unmanned aerial vehicles multispectral imagery

“…The results revealed that repRVI performed best in grain yield assessment at the LGF stage, which is consistent with previous studies ( Hassan et al., 2019 ; Fei et al., 2021a ; Fei et al., 2021b ; Ganeva et al., 2022 ), because the LGF stage is close to maturity and the information in the UAVs field of view is mainly provided by the mature spikes. The signal is minimally affected by moisture and other green parts of the rice plant.…”

Research hotspots and frontiers in agricultural multispectral technology: Bibliometrics and scientometrics analysis of the Web of Science

“…Keywords clustering 4 biomass: Biomass is a common crop parameter based on remote sensing and with the rapid development of remote sensing technology biomass detection techniques have advanced tremendously with the rapid development of precision agriculture from 1980 to 2021. The rapid development of UAV technology, lightweight multispectral, and hyperspectral equipment has provided new tools for biomass detection and during these decades most of the studies on crop parameters were conducted based on spectral information and with the addition of 3D information technology the interplay of the two is a new progress in the detection of crop parameters ( Candiago et al, 2015 ; Näsi et al, 2018 ; Zhu et al, 2019a ; Fei et al, 2021 ; Jayakumari et al, 2021 ; Li et al, 2021 ; Yu et al, 2021 ).…”

Soybean leaf estimation based on RGB images and machine learning methods

“…Precision agriculture has become a popular topic in recent years, and the development of nondestructive estimation technologies have provided new methods and means for crop growth estimation, presenting good application prospects. Previous studies have shown that images collected by sensors such as an RGB camera [ 17 ], a thermal infrared camera [ 18 ], a hyperspectral camera [ 19 ], and a CT scanner [ 20 ] can extract multiple image traits, and based on these image traits, prediction models of the leaf area index [ 21 ], leaf iron deficiency greening, and other indicators [ 22 , 23 ] can be established. Among the above-mentioned sensors, thermal infrared cameras work in the field environment, which is greatly affected by the ambient temperature and has an extremely low resolution [ 24 ].…”