EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Full-Spectrum Hyperspectral Modeling of Leaf Dry Matter Content Using a Stacked Ensemble Framework

Reinis Alksnis iD, Ina Alsina iD, Mara Duma iD, Laila Dubova iD, Uldis Gross iD, Tetiana Harbovska iD

Sensors · 06.03.2026 · 26 (5) · 1665

DOI: 10.3390/s26051665 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 06.03.2026. Verfasst von Reinis Alksnis und 5 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Engineering and Information Technology, Latvia University of Life Sciences and Technologies, LV-3001 Jelgava, Latvia. Hauptthemen laut MEDLINE: Plant Leaves, Hyperspectral Imaging, Spectrum Analysis. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: LBTU institucionālās kapacitātes stiprināšana izcilībai studijās un pētniecībā (ANM1), The Recovery and Resilience Facility.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The objective of this study was to assess the predictability of leaf dry matter content across a diverse range of plant species using hyperspectral reflectance data. The dataset encompassed leaves from multiple crops, including potatoes, beans, wheat, maize, peas, tomatoes, basil, and cucumbers, collected under varying growth conditions, cultivation systems, seasonal contexts, and developmental stages. As an initial benchmark, commonly used narrow-band spectral indices and their combinations were evaluated, but they exhibited limited predictive performance for dry matter content. Consequently, several full-spectrum machine learning models were trained and compared to assess their individual predictive ability. Given their complementary strengths, these models were integrated into a stacked ensemble framework to enhance overall accuracy. The resulting ensemble, combining the outputs of multiple base learners through a meta-learner, achieved a coefficient of determination of R2=0.896 on an independent test set, outperforming all individual models. The findings highlight the potential of a multi-model stacking approach to improve the accuracy and robustness of leaf biochemical property estimation from hyperspectral data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Plant LeavesHyperspectral ImagingSpectrum AnalysisCrops, AgriculturalMachine LearningEnsemble Learning

Stichwörter der Autor:innen

cropsdry matterlight reflectance spectrastacking modelswater indices

Autor:innen und Institutionen

  • Faculty of Engineering and Information Technology, Latvia University of Life Sciences and Technologies, LV-3001 Jelgava, Latvia.
  • Faculty of Agriculture and Food Technologies, Latvia University of Life Sciences and Technologies, LV-3001 Jelgava, Latvia.

ORCID-Profile: Reinis Alksnis, Ina Alsina, Mara Duma, Laila Dubova, Uldis Gross, Tetiana Harbovska

Förderung

  • LBTU institucionālās kapacitātes stiprināšana izcilībai studijās un pētniecībā (ANM1)
  • The Recovery and Resilience Facility

Zitieren

APA

Alksnis, R., Alsina, I., Duma, M., Dubova, L., Gross, U., & Harbovska, T. (2026). Full-Spectrum Hyperspectral Modeling of Leaf Dry Matter Content Using a Stacked Ensemble Framework. Sensors, 26(5), 1665. https://doi.org/10.3390/s26051665

Vancouver

Alksnis R, Alsina I, Duma M, Dubova L, Gross U, Harbovska T. Full-Spectrum Hyperspectral Modeling of Leaf Dry Matter Content Using a Stacked Ensemble Framework. Sensors. 2026;26(5):1665. doi:10.3390/s26051665

BibTeX
@article{Alksnis2026,
  author = {Alksnis, Reinis and Alsina, Ina and Duma, Mara and Dubova, Laila and Gross, Uldis and Harbovska, Tetiana},
  title = {Full-Spectrum Hyperspectral Modeling of Leaf Dry Matter Content Using a Stacked Ensemble Framework},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {5},
  pages = {1665},
  doi = {10.3390/s26051665}
}
Kostenlos starten →