EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Scalable thermal imaging and processing framework for water status monitoring in vineyards

André Pires, Alexandre Bernardino, Gonçalo Victorino, J. Miguel Costa, Carlos M. Lopes, José Santos-Victor

Computers and Electronics in Agriculture · 2025-12 · 239 · 110931

DOI: 10.1016/j.compag.2025.110931 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computers and Electronics in Agriculture, 2025-12. Verfasst von André Pires und 5 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Viticulture faces growing challenges due to climate change, such as rising temperatures and reduced rainfall during critical growth stages of grapevines. Traditional methods to monitor vine water status are labor-intensive and impractical for large-scale field application, highlighting the need for innovative approaches. This article explores the use of an innovative approach to thermal imaging to automatically estimate stem water potential (Ψstem) in vineyards. Our method combines angled thermal imaging, deep learning-based automatic canopy temperature estimation, and meteorological data to enhance accuracy across diverse conditions. Thermal images were collected from different white grape varieties, at different times of the day and sides of the canopy. Results indicate that the developed thermography-based models using the proposed automatic approach can ex- plain a high and significant percentage of Ψstem variability (Test R2 of 0.598). The best results were obtained when thermal images were taken either at midday or afternoon. Interestingly, model performance remained stable despite variations in weather station proximity and data frequency. This automatic approach to thermography proved to be a viable alternative to soil water probes, demonstrating comparable metrics in this study and showing promise in scenarios where probe coverage is limited. These findings highlight the potential of automatic thermography-based machine learning models as a scalable solution for water status monitoring in vineyards.

Zitieren

APA

Pires, A., Bernardino, A., Victorino, G., Costa, J. M., Lopes, C. M., & Santos-Victor, J. (2025). Scalable thermal imaging and processing framework for water status monitoring in vineyards. Computers and Electronics in Agriculture, 239, 110931. https://doi.org/10.1016/j.compag.2025.110931

Vancouver

Pires A, Bernardino A, Victorino G, Costa JM, Lopes CM, Santos-Victor J. Scalable thermal imaging and processing framework for water status monitoring in vineyards. Computers and Electronics in Agriculture. 2025;239:110931. doi:10.1016/j.compag.2025.110931

BibTeX
@article{Pires2025,
  author = {Pires, André and Bernardino, Alexandre and Victorino, Gonçalo and Costa, J. Miguel and Lopes, Carlos M. and Santos-Victor, José},
  title = {Scalable thermal imaging and processing framework for water status monitoring in vineyards},
  journal = {Computers and Electronics in Agriculture},
  year = {2025},
  volume = {239},
  pages = {110931},
  doi = {10.1016/j.compag.2025.110931}
}
Kostenlos starten →