Researchers from the Demining Research Community have come up with a new and promising technology that focuses on detecting and diffusing unexploded mines and munitions with 92% accuracy. The technology has been tested successfully by the researchers, and now they are creating a landscape in Oklahoma to direct their newly developed “drone-based, machine learning-powered detection system” by deploying grids and mines. As reported by Scientific American magazine, the co-founders of this non-profit organization are leading the research team, and the practice is aimed at removing all those hidden munitions from the ground that could otherwise pose a threat to civilians.
You might be aware of the fact that countries usually deploy these mines and munitions during wartime in order to restrict the enemy from entering their area. This is considered a safe practice during the war, but it has some repercussions to follow. When the war ends, there will remain some unexploded cluster munitions engraved in the ground because they can’t be automatically “turned off”. So, whenever someone walks on them, they could explode, thereby posing a threat to human life. In view of this, this state-of-the-art technology has been developed so that the lives of civilians can be protected.
And the idea can be successfully implemented in Ukraine as, according to some reports, around 160,000 km of the Ukrainian territory contains these “landmines”. According to the researchers, “It’s not easy to find places where you can lay dormant mines and fly drones. We’re trying to automate the detection of different types of landmines, anti-personnel mines, anti-tank mines, and we’re doing this through remote sensing and drones.”
“Currently, most humanitarian demining is manual, so this is a beep and prod method where we have metal detectors, layout grids, and you’re able to scan a metal detector really low down to the ground and poke it to see if there is a mine there,” they further stated. However, Gabriel Steinberg, who is one of the co-founders of the Demining Research Community, said, “Convolutional neural networks are scarily accurate and have a lot of applications that are better than humans. We’ve achieved 92 percent accuracy with the limited amount of data we’ve gotten in this field season.”
The system has got a whopping 92% accuracy, which means that it could really prove useful not only for Ukraine but for other countries fighting wars as well. Considering this, the US announced a package of $89 million on Thursday for Ukraine so that it can acquire this cutting-edge technology for future prospects and be on par with Russia to some extent in combat support.
Steinberg said, “Machine learning always works best with real-world data. We have an algorithm that’s able to detect the PFM-1 (butterfly mine) plastic landmine that’s actually being used and deployed in Ukraine. So, we’re flying over, taking surveys of minefields, and then we’re using machine learning to help us automate the detection afterward. 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.”