A team of researchers from the Demining Research Community is working to automate the detection of different types of landmines, anti-personnel mines and anti-tank mines.

The nonprofit organization is currently setting up grids of mines and munitions in Oklahoma to train a drone-based, machine-learning-powered detection system designed to autonomously locate explosives without risking the lives of human workers manually seeking out these explosives.

The detection technology is currently being trialed at the explosive ordnance disposal field operated by Oklahoma State University where researchers using machine learning algorithms, remote sensing and drones to detect landmines have achieved a reported 92% rate of accuracy from the data accrued from the field so far.

“The goal would be to fly over with different sensors, either a thermal sensor, multispectral or visual input the imagery into the algorithm, and it'll output the coordinates of where different mines lay. And we'll also say what type of mine they are," the researchers explained.

With an estimate of millions of active mines and munitions scattered across dozens of countries, researchers from the organization are hoping to make the drone-detection system available to demining organizations across the globe.

For more information, watch the accompanying video that appears courtesy of Scientific America.