top of page

Your certificate is now private

CertificateBackground.png

Certificate of Completion

ErdosHorizontal.png

THIS ACKNOWLEDGES THAT

HAS COMPLETED THE FALL 2026 DEEP LEARNING BOOT CAMP

Jay Desai

Roman Holowinsky, PhD

November 15, 2026

DIRECTOR

DATE

clear.png

TEAM

Spotting Invasive Plant Species from Citizen-Science Photos

Rachel Neville, Craig Corsi, Apostolos Zournas, Jay Desai

clear.png

Invasive plant species adversely affect native ecosystems, out-competing native species for resources such as water and soil nutrients. Invasive species fundamentally change the ecosystem. In the Southwest, this impacts fire resistance, vital biodiversity, and food webs. Early detection of small and localized infestations before the need for large-scale or expensive interventions can be a vital tool for containment. This project builds an image classifier that distinguishes regionally invasive plant species from visually similar native species, targeting invasive species of the southwest US. We curated a dataset from the research-grade citizen-science photos from the iNaturalist API. We fine tuned DINOv3 vision transformer to identify 40 invasive species with an 93.1% top-1 accuracy.

Screen Shot 2022-06-03 at 11.31.35 AM.png
github URL

©2017-2026 by The Erdős Institute.

bottom of page