Detecting cow behaviours associated with parturition using computer visionTools McDonagh, John (2023) Detecting cow behaviours associated with parturition using computer vision. PhD thesis, University of Nottingham.
AbstractMonitoring of dairy cows and their calf during parturition is essential in determining if there are any associated problems for mother and offspring. This is a critical period in the productive life of the mother and offspring. A difficult and assisted calving can impact on the subsequent milk production, health and fertility of a cow, and its potential survival. Furthermore, an alert to the need for any assistance would enhance animal and stockperson wellbeing. Manual monitoring of animal behaviour from images has been used for decades, but is very labour intensive. Recent technological advances in the field of Computer Vision based on the technique of Deep Learning have emerged, which now makes automated monitoring of surveillance video feeds feasible. The benefits of using image analysis compared to other monitoring systems is that image analysis relies upon neither transponder attachments, nor invasive tools and may provide more information at a relatively low cost. Image analysis can also detect and track the calf, which is not possible using other monitoring methods. Using cameras to monitor animals is commonly used, however, automated detection of behaviours is new especially for livestock.
Actions (Archive Staff Only)
|
Tools
Tools
