Italian researchers have found that using AI-equipped drones and high-tech sensor systems could help detect pest infestations in large orchards, allowing farmers to pinpoint areas where pest control is needed more easily.
Global crop losses from pests and diseases are a huge burden on the economy, farmers, and consumers. According to the U.S. Department of Agriculture, pests destroy 20-40% of crops worldwide each year, costing the economy around $220 billion.
In 2019, the brown marmorated stink bug caused nearly $640 million in damage to crops in Italy alone, according to Interesting Engineering.
While farmers typically use pheromone traps, sweep netting, and visual monitoring to eradicate this nuisance bug, these methods are time-consuming and challenging to implement when dealing with sprawling orchards.
However, researchers from the University of Modena and Reggio Emilia in Italy believe AI-enabled drones could be the perfect solution, as Interesting Engineering reported.
The team, led by Lara Maistrello, a life sciences associate professor at the University of Modena, programmed a specific flight path for the drones so they would fly 26 feet above pear orchards to capture images of potential pest infestations. The cameras took high-resolution photos of the orchards, which were used to train AI models in detecting pest invasions.
"Models trained using this data were more effective at spotting the stink bug with an accuracy of 97% compared to those trained from scratch," Interesting Engineering explained.
Researchers said this method could be used for other aspects of pest management, including accurate forecasting that can be updated as environmental conditions change.
In another study, engineers from the City University of New York, the University of Melbourne, RMIT University, and the ARC Centre of Excellence for Transformative Meta-Optical Systems developed a compact crop-health sensor that could further efforts to identify pests without the need for human labor.
According to RMIT Australia, the flat-optics sensor can alternate quickly between detailed infrared imaging and edge detection (capturing the outline of something, such as a pear). The engineering team says it can be easily integrated into drones for remote crop monitoring, allowing farmers to keep tabs on which crops need irrigation, fertilizers, and pest control simultaneously.
In turn, this would theoretically result in better harvests, helping farmers reduce food waste and lowering grocery costs.
Farmers are already using AI-guided drones to spray herbicides and pesticides, and the team of engineers is taking steps to bring its sensor system to the industries that need it. RMIT holds a U.S. patent and has an Australian patent pending for its system of producing vanadium dioxide films, a key component of the smart sensor, and the researchers believe it can be seamlessly incorporated into existing manufacturing systems.
From heat-resistant plants to a soil-steaming machine that can kill troublesome crop pests, scientists are scrambling to come up with creative ways to protect our food supply from rising temperatures, diseases, bugs, and pollution, all of which are made worse by the changing climate. However, by wielding the power of technology, the agricultural industry has plenty of tools at its disposal to adjust to the climate.
Join our free newsletter for weekly updates on the coolest innovations improving our lives and saving our planet.
