Researchers at the University of California, Riverside, are developing an automated device to replace sticky traps growers commonly use to monitor insects.

In trials, scientists used a sensor that consists of a phototransistor array connected to an electronic board and a laser pointing at the phototransistor array. Light fluctuations caused by insects crossing the beam were captured. The data was fed into a digital sound recorder and recorded as an MP3 and downloaded to a computer. Data was then incorporated into classification algorithms.

The university’s goal is to make this automated classification as simple as current methods, such as sticky traps and interception traps, but with the advantages of greater accuracy, real-time monitoring, and the ability to collect additional flight behavior patterns.

Dr. Eamonn Keogh, UCR computer science professor who headed the research team, built the original sensor with plastic Lego building blocks, a 99-cent off-the-shelf laser pointer, and a part from a television remote control. He believes the sensors can be built for less than $10 and powered by a solar energy or long-lasting battery.

The UC scientists worked with six insect species and collected data for three years. As they added additional insect flight behavior patterns to their classification algorithm, the accuracy of insect classification increased.

For example, using only wing beat sounds gave an 88 percent identification success rate. But adding time of day increased the success rate to 95 percent, and adding location further increased success to 97 percent. Researchers believe that adding variables, like the height at which insects fly and the temperature and humidity of the environment, will further improve success. •