Deep learning-based estimation of plant functional traits from canopy spectral reflectance
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy · 2026-10 · 359 · 127895
DOI: 10.1016/j.saa.2026.127895 ↗Kurz gefasst
Fachartikel in der Zeitschrift Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2026-10. Verfasst von Wenchao Qi und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Earth System Science, Ministry of Education Key Laboratory for Earth System Modeling, Institute for Global Change Studies, Tsinghua Uni…. Hauptthemen laut MEDLINE: Deep Learning, Spectrum Analysis, Plant Leaves, Plants. Die Arbeit nennt 58 Literaturangaben. Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China.
Abstract (Originalsprache)
Large-scale canopy-level plant trait quantification enhances crop yield and quality assessment, supports sustainable forestry economic development, and improves ecosystem monitoring globally. However, traditional methods relying on vegetation spectral libraries and machine learning models often face challenges in capturing the nonlinear and multivariate characteristics of canopy spectral responses. To overcome these challenges, we propose a novel deep learning framework, the Canopy-level Plant Functional Trait Retrieval Network (CTRN), which integrates Kolmogorov-Arnold Networks (KAN), Transformer, and Convolutional Neural Networks (CNN) to effectively extract informative representations from high-dimensional hyperspectral reflectance. The model is trained and evaluated using a comprehensive spectral-trait dataset, covering various plant species, different sensors, and multiple continents, and focuses on ten key canopy functional traits. Experimental results show that CTRN consistently outperforms other models, achieving R2 values greater than 0.82 across all traits. Furthermore, even with just 44 spectral bands at a 40 nm resolution, CTRN demonstrates commendable accuracy in estimating LMA and C, with R2 values approaching 0.80. These findings highlight the robust ability of the model to characterize complex associations between canopy spectra and plant functional traits, supporting accurate parameter retrieval in ecological and agricultural applications.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Earth System Science, Ministry of Education Key Laboratory for Earth System Modeling, Institute for Global Change Studies, Tsinghua University, Beijing 100084, China.
- Department of Earth System Science, Ministry of Education Key Laboratory for Earth System Modeling, Institute for Global Change Studies, Tsinghua University, Beijing 100084, China; Ministry of Education Ecological Field…
- Institute of Aerospace Remote Sensing Innovations, Guangzhou University, Guangzhou, Guangdong 510006, China. Electronic address: wangyb@gzhu.edu.cn.
- Northeast Asia Biodiversity Research Center, College of Forestry, Northeast Forestry University, Harbin 150040, China.
Förderung
- National Natural Science Foundation of China
- National Key Research and Development Program of China
Zitieren
APA
Qi, W., Yu, L., Wang, Y., Liu, T., Zhao, Q., & Wu, H. (2026). Deep learning-based estimation of plant functional traits from canopy spectral reflectance. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 359, 127895. https://doi.org/10.1016/j.saa.2026.127895
Vancouver
Qi W, Yu L, Wang Y, Liu T, Zhao Q, Wu H. Deep learning-based estimation of plant functional traits from canopy spectral reflectance. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 2026;359:127895. doi:10.1016/j.saa.2026.127895
BibTeX
@article{Qi2026,
author = {Qi, Wenchao and Yu, Le and Wang, Yibo and Liu, Tao and Zhao, Qiang and Wu, Hui},
title = {Deep learning-based estimation of plant functional traits from canopy spectral reflectance},
journal = {Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy},
year = {2026},
volume = {359},
pages = {127895},
doi = {10.1016/j.saa.2026.127895}
}
EUVIMED