Drones and AI: The Winning Combination for Modern Agriculture

Thursday 20 June 2024 - 14:00

Agriculture, like many other sectors, is embracing modernization and technological advancements to address its growing challenges. Faced with the increasing threat of pests and the imperative to optimize yields, some farmers are turning to cutting-edge technology. Today, the combined use of drones and artificial intelligence (AI) stands out as one of the most promising alliances for crop monitoring. This technological synergy not only enables proactive problem detection but also significantly enhances agricultural management, making the entire process more precise and responsive.

The Stink Bug: An Agricultural Scourge

Known for their unpleasant odor when crushed, stink bugs are actually one of the most formidable agricultural pests in North America and Southern Europe. Specifically, the brown marmorated stink bug (Halyomorpha halys), originally from Asia, caused an estimated €500 million in damage across Europe in 2019 alone. These pests target apple trees, pear trees, vineyards, soybeans, and corn, posing a severe threat to a wide range of crops.

Monitoring these pests is crucial for the sustainability of crops. However, current methods are often labor-intensive and inefficient on a large scale. A recent study by Italian researchers, published in the April issue of *Pest Management Science*, explored the use of drones equipped with cameras and AI algorithms to monitor stink bugs in pear orchards.

Agriculture 3.0: A New Era of Crop Surveillance

For their experiment, the scientists utilized a commercial drone, the DJI Matrice 300, equipped with a high-definition camera capable of capturing detailed images of the infested orchard. By developing an automated flight protocol, they remotely piloted the drone via a mobile application, allowing it to survey the areas affected by the insect.

The drone captured hundreds of high-quality images, which were then used to train an AI algorithm capable of identifying stink bugs with 97% accuracy. Daniele Giannetti, a researcher at the University of Parma and co-author of the study, stated, "This innovative surveillance system has demonstrated the potential of integrating drones and AI to detect and quantify pest presence."

Outside of this experiment, monitoring stink bugs is an extremely arduous task, typically involving the use of pheromone traps to attract the insects, followed by manual counting. While this method is widespread, labor costs are very high, and its effectiveness is limited compared to the drone-AI method.

During the drone flights, the researchers observed an unexpected behavior from the stink bugs. Instead of fleeing or falling off the plants, the pests remained perfectly still as the drone flew overhead at altitudes between 4 and 8 meters. Approximately 85% of the observed specimens exhibited this freezing behavior, allowing the drone to capture exceptionally clear images with the bugs clearly visible in the frame. Of the 2,459 images taken by the drone, 402 contained stink bugs.

The researchers then manually annotated these images to train the AI algorithm, which proved extremely effective in identifying the pests in other images. Although the study focused on stink bugs, the scientists believe that the same principles could be applied to monitoring other pests by using drone images to train AI models specialized in this task.

Exciting Prospects for Agricultural Management

The integration of drones and AI in agriculture opens up exciting prospects for crop management and pest control. Lara Maistrello, a professor at the University of Modena and co-author of the study, expressed particular enthusiasm: "This experiment is truly promising. We welcome these results with great interest, especially due to the wide range of future applications." Indeed, other potential applications are equally enticing: soil analysis, monitoring weed spread, optimizing irrigation, and more.

Italian researchers have developed a novel protocol to monitor a pear orchard infested with stink bugs. They used a drone equipped with a camera and fed with AI algorithms to detect the bugs in the trees. This method is far more effective than traditional methods and could be applied to other areas of the agricultural sector.

In a world where precision and efficiency are paramount, the fusion of drone technology and AI heralds a new era of agricultural innovation, promising to transform how we manage and protect our crops.