Phenotyping agronomic and physiological traits in peanut under mid‐season drought stress using UAV‐based hyperspectral imaging and machine learning

Abstract

Agronomic and physiological traits in peanut (Arachis hypogaea) are important to breeders for selecting high‐yielding and resilient genotypes. However, direct measurement of these traits is labor‐intensive and time‐consuming. This study assessed the feasibility of using unmanned aerial vehicles (UAV)‐based hyperspectral imaging and machine learning (ML) techniques to predict three agronomic traits (biomass, pod count, and yield) and two physiological traits (photosynthesis and stomatal conductance) in peanut under drought stress. Two different approaches were evaluated. The first approach employed eighty narrowband vegetation indices as input features for an ensemble model that included K‐nearest neighbors, support vector regression, random forest, and multi‐layer perceptron (MLP). The second approach utilized mean and standard deviation of canopy spectral reflectance per band. The resultant 400 features were used to train a deep learning (DL) model consisting of one‐dimensional convolutional layers followed by an MLP regressor. Predictions of the agronomic traits obtained using feature learning and DL (R2 = 0.45–0.73; symmetric mean absolute percentage error [sMAPE] = 24%–51%) outperformed those obtained using feature engineering and conventional ML models (R2 = 0.44–0.61, sMAPE = 27%–59%). In contrast, the ensemble model had a slightly better performance in predicting physiological traits (R2 = 0.35–0.57; sMAPE = 37%–70%) compared to the results obtained from the DL model (R2 = 0.36–0.52; sMAPE = 47%–64%). The results showed that the combination of UAV‐based hyperspectral imaging and ML techniques have the potential to assist breeders in rapid screening of genotypes for improved yield and drought tolerance in peanut.

Yield prediction in a peanut breeding program using remote sensing data and machine learning algorithms

“…Phenotypes extracted at an elementary level from field plots in this study and others have been directly correlated with yield, but none of these showed a strong or consistent enough correlation to be reliably predictive on their own, indicating the need for more sophisticated ML methodologies ( Manley et al., 2023 ). Across the entire dataset, the performance of the RF and XGBoost models constructed in this study was superior to those seen in previous studies in peanut ( Balota and Oakes, 2016 ; Bagherian et al., 2023 ; Shahi et al., 2023 ). In Balota and Oakes (2016) , the R 2 for yield was significantly lower than in the present study, ranging from 0.26 – 0.39.…”

“…In Balota and Oakes (2016) , the R 2 for yield was significantly lower than in the present study, ranging from 0.26 – 0.39. In Bagherian et al. (2023) , the highest R 2 achieved for estimating yield with Deep Learning (DL) and ML models was 0.61.…”

“…EXtreme Gradient Boosting is an advanced implementation of gradient boosting algorithms that builds trees sequentially, so that each new tree can correct the errors made by each previous tree, resulting in reduced error ( Chen and Guestrin, 2016 ). Recent research in peanuts has shown that ML models can potentially be used to estimate various important traits via the use of the values of visual bands in RGB and multispectral imagery ( Bagherian et al., 2023 ; Shahi et al., 2023 ). However, AI powered by remote sensing data has not yet been used across multiple years and a diverse set of material to assess models for their robustness in peanut ( Bagherian et al., 2023 ; Shahi et al., 2023 ).…”

“…Recent research in peanuts has shown that ML models can potentially be used to estimate various important traits via the use of the values of visual bands in RGB and multispectral imagery ( Bagherian et al., 2023 ; Shahi et al., 2023 ). However, AI powered by remote sensing data has not yet been used across multiple years and a diverse set of material to assess models for their robustness in peanut ( Bagherian et al., 2023 ; Shahi et al., 2023 ). Therefore, it is imperative that ML models to predict yield in peanut are built that are highly robust and remain applicable to unseen data; that is, data that have not yet been encountered by the models ( Yoosefzadeh Najafabadi et al., 2023 ).…”

Applications of Artificial Intelligence, Machine Learning, and Deep Learning in Plant Breeding

“…EXtreme Gradient Boosting is an advanced implementation of gradient boosting algorithms that builds trees sequentially, so that each new tree can correct the errors made by each previous tree, resulting in reduced error . Recent research in peanuts has shown that ML models can potentially be used to estimate various important traits via the use of the values of visual bands in RGB and multispectral imagery (Bagherian et al, 2023;Shahi et al, 2023). However, AI powered by remote sensing data has not yet been used across multiple years and a diverse set of material to assess models for their robustness in peanut (Bagherian et al, 2023;Shahi et al, 2023).…”