AUTHOR=Ting To-Chia , Souza Augusto C. M. , Imel Rachel K. , Guadagno Carmela R. , Hoagland Chris , Yang Yang , Wang Diane R. 

TITLE=Quantifying physiological trait variation with automated hyperspectral imaging in rice

JOURNAL=Frontiers in Plant Science

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2023.1229161

DOI=10.3389/fpls.2023.1229161

ISSN=1664-462X

ABSTRACT=Advancements in hyperspectral imaging (HSI) together with the establishment of dedicated plant

phenotyping facilities worldwide have enabled high-throughput collection of plant spectral images

with the aim of inferring target phenotypes. Here, we test the utility of HSI-derived canopy data,

which were collected as part of an automated plant phenotyping system, to predict physiological

traits in cultivated Asian rice (Oryza sativa). We evaluated 23 genetically diverse rice accessions

from two subpopulations under two contrasting nitrogen conditions and measured 14 leaf- and

canopy-level parameters to serve as ground-reference observations. HSI-derived data were

 used to (1) classify treatment groups across multiple vegetative stages using support vector

 machines (≥ 83% accuracy) and (2) predict leaf-level nitrogen content (N, %, n=88) and carbon

 to nitrogen ratio (C:N, n=88) with Partial Least Squares Regression (PLSR) following RReliefF

 wavelength selection (validation: R2 = 0.797 and RMSEP = 0.264 for N; R2 = 0.592 and RMSEP

 = 1.688 for C:N). Results demonstrated that models developed using training data from one rice

 subpopulation were able to predict N and C:N in the other subpopulation, while models trained

 on a single treatment group were not able to predict samples from the other treatment. Finally,

 optimization of PLSR-RReliefF hyperparameters showed that 300-400 wavelengths generally

 yielded the best model performance with a minimum calibration sample size of 62. Results

 support the use of canopy-level hyperspectral imaging data to estimate leaf-level N and C:N

 across diverse rice, and this work highlights the importance of considering calibration set design

 prior to data collection as well as hyperparameter optimization for model development in future

 studies.