EUVIMEDEUVIMED entdecken
Crossref

Thermal Imaging Based Assessment of Livestock Heat Stress for Climate Resilient Farm Management

Zengling Wang

Journal of Smart and Sustainable Farming · 30.06.2026 · 139

DOI: 10.64026/jssf/2026012 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Smart and Sustainable Farming, online erschienen am 30.06.2026. Verfasst von Zengling Wang. Die Arbeit nennt 16 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the rising temperatures and humidity, the frequency and intensity of heat stress is affecting the productivity, animal welfare and profitability of farms. Environmental conditions are traditionally described by the temperature–humidity index (THI) and this index do not fully describe the physiological responses of the individual. Rectal thermometry, on the other hand, is a good reference measurement, but it is invasive, time consuming, and not suitable for continuous monitoring of a herd. The present study aims to propose a framework for infrared thermography (IRT) for non-invasive assessment of heat stresses in dairy cattle based on radiometric correction, automated body segmentation, region-of-interest (ROI) detection, thermal feature extraction, and machine-learning classification, and to integrate the barn microclimate measurements into this framework. The framework was tested with 120 animal-sessions ranging between THI 65.8 to 89.6. The overall performance of Otsu based Segmentation was good in extracting the animal regions, with an intersection-over-union (IoU) score of 0.95. The maximum eye temperature had a high correlation with the rectal temperature (r = 0.92) and RMSE of 0.15 °C. The logistic regression classifier that integrated eye temperature and THI had an accuracy of 0.84 and an area under the receiver operating characteristic curve (AUC) of 0.92 to identify heat stressed animals based on the criterion of rectal temperature ≥ 39.2 °C. The benefit of adding animal level thermal information was demonstrated by comparing with THI-only classification that had an AUC of 0.75. In addition, a 24-hour monitoring system detected an average delay of 2.25 hours between changes in THI and change in herd eye-temperature. This temporal relationship can be used to make pre-emptive cooling decisions. The proposed framework provides multiple levels of alerts for shade, sprinkler and ventilation management, providing a practical way to improve climate resilient livestock production, animal welfare and even resource utilisation.

Zitieren

APA

Wang, Z. (2026). Thermal Imaging Based Assessment of Livestock Heat Stress for Climate Resilient Farm Management. Journal of Smart and Sustainable Farming, 139. https://doi.org/10.64026/jssf/2026012

Vancouver

Wang Z. Thermal Imaging Based Assessment of Livestock Heat Stress for Climate Resilient Farm Management. Journal of Smart and Sustainable Farming. 2026:139. doi:10.64026/jssf/2026012

BibTeX
@article{Wang2026,
  author = {Wang, Zengling},
  title = {Thermal Imaging Based Assessment of Livestock Heat Stress for Climate Resilient Farm Management},
  journal = {Journal of Smart and Sustainable Farming},
  year = {2026},
  pages = {139},
  doi = {10.64026/jssf/2026012}
}
Kostenlos starten →