This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
High-Performance Automated Detection of Sheep Binocular Eye Temperatures and Their Correlation with Rectal Temperature
by
, ,
Yadan Zhang
1,2
,
Ying Han
1,2,
Xiaocong Li
1,2,
Xueting Zeng
1,2,
Waleid Mohamed EL-Sayed Shakweer
3, 
Gang Liu
1,2,* and 
Jun Wang
4,* 1
Key Laboratory of Smart Agriculture System Integration, Ministry of Education, China Agricultural University, Beijing 100083, China
2
Key Laboratory of Agricultural Information Acquisition Technology, Ministry of Agriculture and Rural Affairs, China Agricultural University, Beijing 100083, China
3
Animal Production Department, Agricultural and Biological Research Institute, National Research Centre, Cairo 12622, Egypt
4
School of Information Engineering, Henan University of Science and Technology, Luoyang 471003, China
*
Authors to whom correspondence should be addressed.
Animals 2025, 15(17), 2475; https://doi.org/10.3390/ani15172475
Submission received: 9 June 2025
/
Revised: 25 July 2025
/
Accepted: 20 August 2025
/
Published: 22 August 2025
(This article belongs to the Special Issue Modern Livestock and Poultry Farming Technologies to Improve Comfort and Animal Welfare Levels)
Simple Summary
Rectal temperature measurement is labor-intensive and induces stress in sheep, potentially compromising both measurement accuracy and animal welfare. Infrared thermography (IRT) offers a non-contact alternative with rapid image acquisition and minimal disturbance. However, most existing research focuses on single-point temperature analysis and often overlooks the diagnostic potential of differences between left and right eye temperatures under varying physiological and environmental conditions. This study integrates deep learning with IRT to detect the left and right eye regions of sheep and extract temperature features. An optimized lightweight model (E-S-YOLO11n) was developed, showing high accuracy and fast detection speed on low-power devices. The temperatures of the two eyes are strongly correlated, but neither exhibits a significant correlation with rectal temperature, which is consistently higher. This suggests that eye temperature is influenced by local environmental factors and may not be a reliable surrogate for rectal temperature. These findings offer practical solutions for precision animal husbandry and support more effective scientific herd management.
Abstract
Although rectal temperature is reliable, its measurement requires manual handling and causes stress to animals. IRT provides a non-contact alternative but often ignores bilateral eye temperature differences. This study presents an E-S-YOLO11n model for the automated detection of the binocular regions of sheep, which achieves remarkable performance with a precision of 98.2%, recall of 98.5%, mAP@0.5 of 99.40%, F1 score of 98.35%, FPS of 322.58 frame/s, parameters of 7.27 M, model size of 3.97 MB, and GFLOPs of 1.38. Right and left eye temperatures exhibit a strong correlation (r = 0.8076, p < 0.0001), However, the eye temperatures show only very weak correlation with rectal temperature (right eye: r = 0.0852; left eye: r = −0.0359), and neither figure reaches statistical significance. Rectal temperature is 7.37% and 7.69% higher than the right and left eye temperatures, respectively. Additionally, the right eye temperature is slightly higher than the left eye (p < 0.01). The study demonstrates the feasibility of combining IRT and deep learning for non-invasive eye temperature monitoring, although environmental factors may limit it as a proxy for rectal temperature. These results support the development of efficient thermal monitoring tools for precision animal husbandry.
