Automatic Monitoring and Population Dynamics Modelling of Microscopic Insect Pest-Predator Interactions in Horticultural Crops

A sustainable approach to pest management is Integrated Pest Management (IPM) which uses bio-control agents. To implement IPM strategies, it is important to accurately identify pests and bio-control predators and understand their population dynamics. This thesis improves current practices by using machine learning and computer vision, coupled with computer simulations, to estimate pest and predator numbers and to model their population dynamics. The downstream benefits of the application of these technologies will be better decisions to enable reduced pesticide use, more effective pest control, more reliable crop yields, and a reduced environmental footprint for horticultural crop production.