Prof Christos Tachtatzis (University of Strathclyde), Prof Gareth Arnott (Queen's University Belfast), Andrew Gardner (Galebreaker Ltd).

Climate change is placing increasing pressure on livestock systems, with rising temperatures and humidity leading to heat stress in cattle. Heat stress reduces animal welfare, lowers productivity, and contributes to economic and environmental inefficiency in food production. Developing tools to detect and mitigate heat stress is therefore crucial for building sustainable and climate-resilient agri-food systems.

This project will develop cutting-edge computer vision approaches to automatically extract respiration rates from calves and cows as a key physiological marker of heat stress. Respiration rate is a sensitive and early indicator of heat load, but existing measurement techniques are invasive, labour-intensive, or impractical at scale. By leveraging video data (including visible and thermal imaging), the student will design algorithms to detect subtle movements or thermal fluctuations associated with breathing.

To provide a holistic assessment, respiration data will be combined with environmental measures such as temperature and humidity, specifically through the Temperature-Humidity Index (THI), a widely used predictor of heat stress. This fusion of animal- and environment-level data will enable robust prediction of heat stress events. A further challenge, and key focus of the PhD, will be the development of reliable methods for individual animal identification and attribution of respiration rates in group-housed environments, using biometric features such as facial or coat pattern recognition.

The student will gain advanced training in computer vision, machine learning, sensor fusion, and precision livestock farming. They will work with multidisciplinary teams and industry stakeholders to validate their system in real farm environments. The outcomes will not only contribute to animal welfare and sustainability but also advance AI methodologies for physiological monitoring in complex, real-world settings.

This project offers the opportunity to contribute to the transformation of livestock farming into a more sustainable, efficient, and climate-resilient system through AI innovation.