UAV-based remote sensing of bee nesting aggregations with computer vision

Pollinating insects are in decline globally, threatening pollination services and driving a growing interest in pollinator monitoring and conservation. However, the implementation of conservation programs for pollinating insects is often complicated by labor-intensive monitoring methods and insufficient data. We detail a rapid, cost-effective solution for surveying and censusing ground nesting bee aggregations, pairing automated UAV image capture with a custom trained computer vision based object detection workflow. To highlight the ease of application and accuracy of this method, we surveyed a Colletes inaequalis nesting aggregation. Our model detected 88% of the nests present in our test dataset with a true positive rate of 97% and returned a total nest count nearly identical to the total estimated from manual counting from imagery. Our model detected nests 20 times faster than the manual counts while mapping the aggregation with sub-millimeter accuracy. Spatial analyses show that bee nest density was heterogeneous, with dense spatially clustered regions.