Machine learning for semi‐automated meteorite recovery. (1st December 2020)

• Record Type: Journal Article
• Authors: Anderson, Seamus; Towner, Martin; Bland, Phil; Haikings, Christopher; Volante, William; Sansom, Eleanor; Devillepoix, Hadrien; Shober, Patrick; Hartig, Benjamin; Cupak, Martin; Jansen‐Sturgeon, Trent; Howie, Robert; Benedix, Gretchen; Deacon, Geoff
• Is Part Of: Meteoritics & planetary science. Volume 55:Number 11(2020)
• Page Start: 2461
• Page End: 2471
• Publication Date: 2020-12-01
• Subjects: Meteorites -- Periodicals; Planetology -- Periodicals; 523.4
• DOI: 10.1111/maps.13593
• Languages: English
• ISSNs: 1086-9379

Abstract

Abstract: We present a novel methodology for recovering meteorite falls observed and constrained by fireball networks, using drones and machine learning algorithms. This approach uses images of the local terrain for a given fall site to train an artificial neural network, designed to detect meteorite candidates. We have field tested our methodology to show a meteorite detection rate between 75% and 97%, while also providing an efficient mechanism to eliminate false positives. Our tests at a number of locations within Western Australia also showcase the ability for this training scheme to generalize a model to learn localized terrain features. Our model training approach was also able to correctly identify three meteorites in their native fall sites that were found using traditional searching techniques. Our methodology will be used to recover meteorite falls in a wide range of locations within globe‐spanning fireball networks.