This OSA/SDPMP project, funded by WILDLABS and Arm, aimed to enhance pollinator monitoring through deep learning. Using Kitware’s VIAME platform, the project involved annotating pollinator videos and training neural networks to automate insect detection and classification. Videos were collected from the San Diego Botanic Garden, focusing on Encinitas baccharis (Baccharis vanessae), an endangered plant species. The project successfully demonstrated that AI can track pollinators in video data, providing a scalable approach to support future conservation efforts.
Pollinator Monitoring Projects
Applying Advanced Tech to Pollinator Monitoring
Stepping "soil-side" in this project effort, OSA provides technical expertise to the San Diego Pollinator Monitoring Program (SDPMP) in their biodiversity monitoring endeavors. Our contributions span digital data collection, biodiversity analysis, deep learning for insect tracking in videos, and ecosystem services for post-fire recovery. This collaboration aims to advance monitoring techniques for critical terrestrial species and habitats.
Learn More About SDPMP Read our Ecology & Evolution Paper
San Diego County Post-Fire Recovery of Pollinators
In this Earth Science Information Partners (ESIP) funded effort, SDPMP with support from OSA evaluated the impact of fire interval and frequency on pollinating insect assemblages in Coastal Sage Scrub habitat of San Diego County. OSA contributed through Ecosystem Services modeling and ArcGIS Dashboard development. The project combined in-situ focal observations and the InVEST crop pollination model to assess the effects of fire, uncovering the importance of reservoir habitat to pollinator diversity.
ESIP Project on GitHub
