Advancing image-based automated trap use for monitoring insects in stored product environments

The stored product environment is generally any part of the post-harvest supply chain, including storage of both raw and processed materials. The variety of commodities, duration of storage, different storage environments, diversity of packaging, and cultural practices all create a system in which monitoring and subsequent prediction of insect pest infestations is difficult. The profit margins within this system require timely monitoring for insect infestations as significant damage to products can result in significant economic losses. However, current monitoring practices can create a delay in the deployment of necessary treatments to protect commodities. The use of artificial intelligence (AI) to aid in monitoring these tiny insect pests may open opportunities for better protection and timely treatments. AI optimizes crop health and yield, reduces health risks from pests, and drives pest management innovation. Additionally, AI ensures sustainable crop protection and aids in eco-friendly pest control. Because storage environments can vary from location to location, development of these AI products must have careful consideration for flexibility and offline applications. Using image-based identification and floorplan mapping, we are beginning to develop an automated monitoring system for real-time planning and tracking in a variety of environments. As researchers, we must create a backbone of monitoring systems using AI driven technology that can constantly adapt to meet user needs, moving from simulations to deployment. However, there is still a need for extensive field testing to ensure the reliability and accuracy of these systems under diverse real-world conditions.