Precision Pollination: A Low-Cost, Scalable Device for Monitoring Pollinators and Pollination Services Using Deep Learning
Insect pollinators are critical to food production and conservation globally. Currently, yields of many crops are limited by insufficient pollination, a problem that declining pollinator populations threaten to accelerate. Despite mounting concern over insect declines, quantifying pollinators remains a significant challenge, in turn limiting our ability to design effective pollinator management strategies. This project will develop a first-of-its-kind device for automatically detecting and identifying insects in the field. These Autonomous Pollinator Sampling units (or AutoPolls) will use cutting-edge deep learning algorithms to quantify pollinator biodiversity and activity on a small, battery-powered device. The project team will use this technology to quantify the ecological diversity of pollinator communities and assess the impacts of practical approaches for improving the climate-resilience of pollination services. The technology that will be developed has the potential to transform the study of pollinators (and other insects), and the critical role they play in supporting crop yields and biodiversity.
James Crall, assistant professor of entomology
Claudio Gratton, professor of entomology
Joshua San Miguel, assistant professor of electrical and computer engineering