Leafy Spurge Dataset: Real-world Weed Classification Within Aerial Drone Imagery

TL;DR

This paper presents a new drone-captured dataset for classifying leafy spurge, an invasive weed, using deep learning, achieving an accuracy of 84%, and aims to aid ecological and remote sensing efforts.

Contribution

The paper introduces a novel aerial imagery dataset for leafy spurge and demonstrates the use of a pre-trained vision transformer for weed classification.

Findings

Achieved 84% accuracy in identifying leafy spurge

Provided a publicly available dataset for machine learning research

Showed that weed classification from drone imagery is feasible

Abstract

Invasive plant species are detrimental to the ecology of both agricultural and wildland areas. Euphorbia esula, or leafy spurge, is one such plant that has spread through much of North America from Eastern Europe. When paired with contemporary computer vision systems, unmanned aerial vehicles, or drones, offer the means to track expansion of problem plants, such as leafy spurge, and improve chances of controlling these weeds. We gathered a dataset of leafy spurge presence and absence in grasslands of western Montana, USA, then surveyed these areas with a commercial drone. We trained image classifiers on these data, and our best performing model, a pre-trained DINOv2 vision transformer, identified leafy spurge with 0.84 accuracy (test set). This result indicates that classification of leafy spurge is tractable, but not solved. We release this unique dataset of labelled and unlabelled, aerial drone imagery for the machine learning community to explore. Improving classification performance of leafy spurge would benefit the fields of ecology, conservation, and remote sensing alike. Code and data are available at our website: leafy-spurge-dataset.github.io.