Butterfly Landmines Mapped by Drones, Machine Learning

Original article was published on Communications of the ACM – Artificial Intelligence

By The Engineer (UK)
June 3, 2020
Comments

Rendering of an inert PFM-1 plastic landmine (butterfly mine) with a U.S. coin for scale.

State University of New York at Binghamton researchers have developed a methodology for using drones and advanced machine learning to detect improvised explosive devices and butterfly landmines.

Credit: Binghamton University

Researchers at the State University of New York at Binghamton have found that drones and advanced machine learning can be used to detect improvised explosive devices (IEDs) and butterfly landmines (surface plastic landmines with low-pressure triggers).

The researchers used convolutional neural networks (CNNs) to develop a method for automating detection and mapping of landmines.

They said a CNN-based approach is much faster than manually counting landmines from an aerial image, and unlike subjective human visual detection, it is quantitative and reproducible.

Binghamton’s Alek Nikulin said drone-assisted mapping and automated detection of scatterable mine fields “would assist in addressing the deadly legacy of widespread use of small scatterable landmines in recent armed conflicts and allow to develop a functional framework to effectively address their possible future use.”

From The Engineer (UK)
View Full Article

 

Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA


 

No entries found