Using AI to Control Pests in Grain Production
Insects can be a real pest to grain producers, literally! Grain producers have to constantly monitor pests when storing grains to ensure postharvest grain quality. However, current sampling and monitoring methods are time-consuming, labor-intensive, and require expertise for accurate species identification. ARS scientists in Manhattan, KS, used deep learning methods and artificial intelligence (AI) to develop image-based identification for five common stored grain insect species: lesser grain borer, rusty grain beetle, red flour beetle, rice weevil, and saw-toothed grain beetle.
The AI-driven system more efficiently identified all species with an accuracy level of at least 96% and enabled producers to more rapidly apply pest controls and ultimately reduce damage and economic losses. This work is part of a broader effort to develop camera-based systems for automated pest monitoring in warehouses, flour mills, and general food storage facilities that will improve pest identification and control.
Related Information
Research Project: Advancing Technologies for Grain Trait Measurement and Storage Preservation