It is widely accepted that urban vegetation can mitigate ambient particulate matter pollution, as plant leaves can capture airborne particles.Many studies have investigated the accumulation of plant leaves on particle retention based on particle mass and/or size distribution.Fewer studies,however, have examined the particle density, and the size and shape characteristics of particles,which may have important implications for source apportionment.Here, we applied an object-based classification approach to automatically extract the particles from scanning electron microscope photos for three plant species.