Communications of the ACM
AI, Drones Will Help Pin Down Sosnovsky's Hogweed
Scientists at the Skolkovo Institute of Science and Technology have developed an artificial intelligence monitoring system that performs real-time image segmentation onboard drones to identify the presence of Sosnovsky's hogweed.
Credit: Skoltech
To help contain the spread of Sosnovsky's hogweed, a plant hazardous to agriculture, local ecosystems, and human health, across Russia, scientists at the Skolkovo Institute of Science and Technology (Skoltech) have developed an artificial intelligence monitoring system that performs real-time image segmentation onboard drones to identify the toxic weed.
The system uses drones that can capture high-resolution images even in cloudy weather via in-flight data acquisition and processing.
Each drone's computer runs heavy segmentation algorithms based on Fully Convolutional Neural Networks (FCNN) that can identify an irregularly shaped object on a pixel-by-pixel basis.
The researchers utilized popular architectures for the FCNN and adapted them for a single-board computer.
Skoltech's Andrey Somov said, "We installed and flight-tested our monitoring system on board the drone which covered an area of up to 28 hectares in 40 minutes, flying at an altitude of 10 meters. And it did not miss a single weed!"
From Skolkovo Institute of Science and Technology (Russia)
View Full Article
Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA
No entries found