High-Performance Automated Detection of Sheep Binocular Eye Temperatures and Their Correlation with Rectal...

Area of Science:

• Animal Science
• Biotechnology
• Computer Vision

Background:

• Rectal temperature measurement in animals is invasive and stressful.
• Infrared thermography (IRT) offers a non-contact alternative but struggles with bilateral temperature differences.
• Automated monitoring is crucial for precision animal husbandry.

Purpose of the Study:

• To develop and evaluate an automated system for detecting sheep eye regions using deep learning.
• To assess the correlation between non-invasively measured eye temperatures and rectal temperatures.
• To explore the potential of IRT combined with deep learning for animal thermal monitoring.

Main Methods:

• Implementation of the E-S-YOLO11n model for automated binocular region detection in sheep.
• Utilizing infrared thermography (IRT) for non-contact temperature measurement.
• Statistical analysis to compare right eye, left eye, and rectal temperatures.

Main Results:

• The E-S-YOLO11n model achieved high performance (precision 98.2%, recall 98.5%, mAP@0.5 99.40%).
• Strong correlation was found between right and left eye temperatures (r = 0.8076).
• Weak and non-significant correlation was observed between eye temperatures and rectal temperature.

Conclusions:

• The E-S-YOLO11n model enables efficient, non-invasive eye temperature monitoring in sheep.
• Eye temperature is not a reliable substitute for rectal temperature due to environmental influences.
• This technology supports advancements in precision animal husbandry tools.