Machine Learning Strategies for the Retrieval of Leaf-Chlorophyll Dynamics: Model Choice, Sequential Versus Retraining Learning, and Hyperspectral Predictors
KAUST DepartmentHydrology, Agriculture and Land Observation Group, Water Desalination and Reuse Center, King Abdullah University of Science and Technology, Thuwal, Saudi Arabia
Environmental Science and Engineering Program
Biological and Environmental Science and Engineering (BESE) Division
Water Desalination and Reuse Research Center (WDRC)
Permanent link to this recordhttp://hdl.handle.net/10754/676461
MetadataShow full item record
AbstractMonitoring leaf Chlorophyll (Chl) in-situ is labor-intensive, limiting representative sampling for detailed mapping of Chl variability at field scales across time. Unmanned aeria-l vehicles (UAV) and hyperspectral cameras provide flexible platforms for observing agricultural systems, overcoming this spatio-temporal sampling constraint. Here, we evaluate a customized machine learning (ML) workflow to retrieve multi-temporal leaf-Chl levels, combining sub-centimeter resolution UAV-hyperspectral imagery (400–1,000 nm) with leaf-level reflectance spectra and SPAD measurements, capturing temporal correlations, selecting relevant predictors, and retrieving accurate results under different conditions. The study is performed within a phenotyping experiment to monitor wild tomato plants’ development. Several analyses were conducted to evaluate multiple ML strategies, including: (1) exploring sequential versus retraining learning; (2) comparing insights gained from using 272 spectral bands versus 60 pigment-based vegetation indices (VIs); and (3) assessing six regression methods (linear, partial-least-square regression; PLSR, decision trees, support vector, ensemble trees, and Gaussian process; GPR). Goodness-of-fit (R2) and accuracy metrics (MAE, RMSE) were determined using training/testing and validation data subsets to assess the models’ performance. Overall, while equally good performance was obtained using either PLSR, GPR, or random forest, results show: (1) the retraining strategy improved the ability of most of the approaches to model SPAD-based Chl dynamics; (2) comparative analysis between retrievals and validation data distributions informed the models’ ability to capture Chl dynamics through SPAD levels; (3) VI predictors slightly improved R2 (e.g., from 0.59 to 0.74 units for GPR) and accuracy (e.g., MAE and RMSE differences of up to 2 SPAD units) in specific algorithms; (4) feature importance examined through these methods, revealed strong overlaps between relevant bands and VI predictors, highlighting a few decisive spectral ranges and indices useful for retrieving leaf-Chl levels. The proposed ML framework allows the retrieval of high-quality spatially distributed and multi-temporal SPAD-based chlorophyll maps at an ultra-high pixel resolution (e.g., 7 mm).
CitationAngel, Y., & McCabe, M. F. (2022). Machine Learning Strategies for the Retrieval of Leaf-Chlorophyll Dynamics: Model Choice, Sequential Versus Retraining Learning, and Hyperspectral Predictors. Frontiers in Plant Science, 13. https://doi.org/10.3389/fpls.2022.722442
SponsorsWe acknowledge Dr. Mitchell Morton, Dr. Sonia Negrao, Gabriele Fiene, and Dinara Utarbayeva from the Salt Lab led by Prof. Mark Tester, with whom field experiments were undertaken. We thank Dr. Kasper Johansen, Dr. Yoann Malbeteau, Samer Al-Mashharawi, Maria Alejandra Perea, Ting Li, Bruno Aragon, Matteo Ziliani, and Cristhian Andrade from the Hydrology, Agriculture and Land Observation (HALO) group, who were also participants of the field data collection. We thank Prof. Magdi A. Mousa and all the workers at the King Abdulaziz University Agricultural Research Station in Hada Al-Sham for their extensive assistance with field maintenance and harvesting.
Funded by the King Abdullah University of Science and Technology (KAUST)
PublisherFrontiers Media SA
JournalFrontiers in Plant Science
PubMed Central IDPMC8963469
Except where otherwise noted, this item's license is described as Archived with thanks to Frontiers in Plant Science under a Creative Commons license, details at: https://creativecommons.org/licenses/by/4.0/
- Machine learning for high-throughput field phenotyping and image processing provides insight into the association of above and below-ground traits in cassava (<i>Manihot esculenta</i> Crantz).
- Authors: Selvaraj MG, Valderrama M, Guzman D, Valencia M, Ruiz H, Acharjee A
- Issue date: 2020
- A robust spectral angle index for remotely assessing soybean canopy chlorophyll content in different growing stages.
- Authors: Yue J, Feng H, Tian Q, Zhou C
- Issue date: 2020
- High throughput analysis of leaf chlorophyll content in sorghum using RGB, hyperspectral, and fluorescence imaging and sensor fusion.
- Authors: Zhang H, Ge Y, Xie X, Atefi A, Wijewardane NK, Thapa S
- Issue date: 2022 May 3
- Disturbed boundaries extraction in coal-grain overlap areas with high groundwater levels using UAV-based visible and multispectral imagery.
- Authors: Guo Y, Zhao Y, Yan H
- Issue date: 2022 Apr 4