Insect Monitoring and Early Detection System in Rice Storage Facilities, 2020

 

Zhongli Pan, adjunct professor, Department of Biological and Agricultural Engineering, UC Davis

The goal of this project is to develop a real-time monitoring and early detection system for insect activity in rice storage facilities. Thus far, engineers have successfully designed, built, tested, and demonstrated a new system that detects insect activity in stored rice with high accuracy, reliability, and low cost and labor. Research in 2020 focused on testing and demonstrating the system in a commercial rice storage facility. Objectives guiding this research included:

• Customize software that has the ability to handle different numbers of traps, along with apps for monitoring the system on both Android and IOS smartphones.

• Scale up the imaging system and demonstrate its effectiveness and accuracy for real-time monitoring and early detection of insect activity in a commercial storage facility.

• Provide recommendations for commercialization of the new system for real-time monitoring and early detection for insect activity in stored rice.

Upgraded imaging system

The imaging system was upgraded to handle many traps. The customized wireless system consisted of 12 insect trap devices, a computer server, website, and the mobile apps.

Researchers installed and tested the system in a commercial rice storage facility at the Sutter Basin Growers Cooperative.
The system was programmed to take images periodically or on demand. Images also could be taken at any time through the computer or the mobile device apps. Data and images were processed in a cloud storage service. The server processed the images and counted the insects captured in each trap with an insect-counting algorithm.

The system also recorded temperature and relative humidity. The user interface displayed images of insects, insect number, temperature, and relative humidity.

Scaling up

Testing of a scaled-up and improved version of the imaging system was delayed because pandemic-related restrictions delayed access to the UC Davis laboratory and workshop where much of this work takes place. Work resumed in June.

Design of the traps and customization of the software was completed in August. The system was installed in a commercial rice storage facility at the the Sutter Basin Growers Cooperative in Knights Landing. Traps were installed in different locations and depths in the storage facility. The signal was lost from the deep traps, so all the traps were ultimately installed in the surface of the stored rice. Three tests were conducted with the rice being inspected by the commercial conventional method before and after installing the new system.

By the end of August, rice had to be moved out of the storage silos to make room for the new crop. To continue monitoring insect activity, rice was loaded into three 20,000-pound metal boxes with four traps installed in each box. Rice was inspected with the conventional method before and after installing the traps. The conventional inspection method and the new system were compared.

Results of the three tests revealed that the new system could detect the first insect within a very short time. For all 12 traps, the average time to detect the first insect ranged from 17 to 33 minutes. Average time for detection of the second insect was 60 to 110 minutes and was 110 to 263 minutes to the third detection.

For all tested traps, the counting accuracy of the new system ranged from 67% to 100%, with an average of 92%. Importantly, no insects were reported using the conventional method.

These results clearly confirm the speed and accuracy of this insect monitoring and early detection system for stored rice. To make the technology widely available to the industry, a full demonstration at different rice storage locations with different storage facilities is needed. Also, the quality and economic benefits of this new technology need to be quantified.