Automatic detection of impact craters on Al foils from the Stardust interstellar dust collector using convolutional neural networks. (26th September 2021)

• Record Type: Journal Article
• Authors: Jaeger, Logan; Butterworth, Anna L.; Gainsforth, Zack; Lettieri, Robert; Zevin, Dan; Ardizzone, Augusto; Capraro, Michael; Burchell, Mark; Wozniakiewicz, Penny; Ogliore, Ryan C.; D Gregorio, Bradley T.; Stroud, Rhonda M.; Westphal, Andrew J.
• Other Names: Brownlee Donald handlingEditor.
• Is Part Of: Meteoritics & planetary science. Volume 56:Number 10(2021)
• Page Start: 1890
• Page End: 1904
• Publication Date: 2021-09-26
• Subjects: Meteorites -- Periodicals; Planetology -- Periodicals; 523.4
• DOI: 10.1111/maps.13747
• Languages: English
• ISSNs: 1086-9379

Abstract

Abstract: NASA's Stardust mission utilized a sample collector composed of aerogel and aluminum foil to return cometary and interstellar particles to Earth. Analysis of the aluminum foil begins with locating craters produced by hypervelocity impacts of cometary and interstellar dust. Interstellar dust craters are typically less than one micrometer in size and are sparsely distributed, making them difficult to find. In this paper, we describe a convolutional neural network based on the VGG16 architecture that achieves high specificity and sensitivity in locating impact craters in the Stardust interstellar collector foils. We evaluate its implications for current and future analyses of Stardust samples.