Researchers Turn to a Drone Equipped with Edge AI Smarts to Spot Invasive Hogweed From the Skies
NVIDIA Jetson Nano strapped to an uncrewed aerial vehicle proves highly capable in finding the invasive weed.
Researchers at the Skolkovo Institute of Science and Technology have turned to machine learning, running in the skies on a drone equipped with an NVIDIA Jetson Nano, to find the invasive hogweed plant — with impressive success.
"The Hogweed of Sosnowskyi (lat. Heracleum sosnówskyi) is poisonous for humans, dangerous for farming crops, and local ecosystems," the researchers explain in the abstract to their paper. "This plant is fast-growing and has already spread all over Eurasia: from Germany to the Siberian part of Russia, and its distribution expands year-by-year."
"In-situ detection of this harmful plant is a tremendous challenge for many countries. Meanwhile, there are no automatic systems for detection and localization of hogweed. In this article, we report on an approach for fast and accurate detection of hogweed. The approach includes the Unmanned Aerial Vehicle (UAV) with an embedded system on board running various Fully Convolutional Neural Networks (FCNN)."
The drone chosen for the task is equipped with an NVIDIA Jetson Nano, a low-cost computer-on-module (COM) designed for low-power high-performance edge AI tasks. Taking a feed from an on-board camera, the system processes the 4k-resolution feed to segment the image into hogweed and not-hogweed — and while it runs at just 0.46 frames per second, it's good enough to detect even individual plants and leaves on-the-wing.
"This system opens up a wide vista for obtaining comprehensive and relevant data about the spreading of harmful plants," the team explains, "allowing for the elimination of their expansion."
The work has been published under closed-access terms in the journal IEEE Transactions on Computers.