Remote sensing estimation of sugar beet SPAD based on un-manned aerial vehicle multispectral imagery
PLOS ONE · 2024
DOI: 10.1371/journal.pone.0300056 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS ONE, 2024. Verfasst von Weishi Gao und 6 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Economic Crops, Xinjiang Academy of Agricultural Sciences, Urumqi, China. Hauptthemen laut MEDLINE: Beta vulgaris, Remote Sensing Technology. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Key Research and development task, National Sugar Industry Technology System Project, Henan Provincial Science and Technology Tackling Project.
Abstract (Originalsprache)
Accurate, non-destructive and cost-effective estimation of crop canopy Soil Plant Analysis De-velopment(SPAD) is crucial for precision agriculture and cultivation management. Unmanned aerial vehicle (UAV) platforms have shown tremendous potential in predicting crop canopy SPAD. This was because they can rapidly and accurately acquire remote sensing spectral data of the crop canopy in real-time. In this study, a UAV equipped with a five-channel multispectral camera (Blue, Green, Red, Red_edge, Nir) was used to acquire multispectral images of sugar beets. These images were then combined with five machine learning models, namely K-Nearest Neighbor, Lasso, Random Forest, RidgeCV and Support Vector Machine (SVM), as well as ground measurement data to predict the canopy SPAD of sugar beets. The results showed that under both normal irrigation and drought stress conditions, the SPAD values in the normal ir-rigation treatment were higher than those in the water-limited treatment. Multiple vegetation indices showed a significant correlation with SPAD, with the highest correlation coefficient reaching 0.60. Among the SPAD prediction models, different models showed high estimation accuracy under both normal irrigation and water-limited conditions. The SVM model demon-strated a good performance with a correlation coefficient (R2) of 0.635, root mean square error (Rmse) of 2.13, and relative error (Re) of 0.80% for the prediction and testing values under normal irrigation. Similarly, for the prediction and testing values under drought stress, the SVM model exhibited a correlation coefficient (R2) of 0.609, root mean square error (Rmse) of 2.71, and rela-tive error (Re) of 0.10%. Overall, the SVM model showed good accuracy and stability in the pre-diction model, greatly facilitating high-throughput phenotyping research of sugar beet canopy SPAD.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Institute of Economic Crops, Xinjiang Academy of Agricultural Sciences, Urumqi, China.
- College of Agronomy, Xinjiang Agricultural University, Urumqi, China.
- Anyang Institute of Technology, AnYang, China.
- Cotton Research Institute, Chinese Academy of Agricultural Sciences, AnYang, China.
ORCID-Profile: Sizhong Li, Jikun Song
Förderung
- Key Research and development task
- National Sugar Industry Technology System Project
- Henan Provincial Science and Technology Tackling Project
Zitieren
APA
Gao, W., Zeng, W., Li, S., Zhang, L., Wang, W., Song, J., & Wu, H. (2024). Remote sensing estimation of sugar beet SPAD based on un-manned aerial vehicle multispectral imagery. PLOS ONE. https://doi.org/10.1371/journal.pone.0300056
Vancouver
Gao W, Zeng W, Li S, Zhang L, Wang W, Song J, et al. Remote sensing estimation of sugar beet SPAD based on un-manned aerial vehicle multispectral imagery. PLOS ONE. 2024. doi:10.1371/journal.pone.0300056
BibTeX
@article{Gao2024,
author = {Gao, Weishi and Zeng, WanYing and Li, Sizhong and Zhang, Liming and Wang, Wei and Song, Jikun and Wu, Hao},
title = {Remote sensing estimation of sugar beet SPAD based on un-manned aerial vehicle multispectral imagery},
journal = {PLOS ONE},
year = {2024},
doi = {10.1371/journal.pone.0300056}
}
EUVIMED