Iron Meteorite Imaging System-ASU
Institution
Arizona State University (Tempe Campus)
Class
Iron Class (2018 – 2019)
Student Team
Daniel Aguiar, Computer Systems Engineering
Katie Baerwaldt, Electrical Engineering
Zakk Giacometti, Computer Systems Engineering
Jacob Hann, Computer Systems Engineering
Andrew Suarez, Engineering Management
Scientific & Technical Guidance
Dr. Laurence Garvie, Research Professor, ASU Center for Meteorite Studies
Dr. Tim McCoy, Curator-in-Charge, US National Meteorite Collection, Smithsonian National Museum of Natural History
Academic Guidance
Dr. Michael Kozicki, Professor, ASU School of Electrical, Computer and Energy Engineering
Anthony Kuhn, Lecturer, ASU Engineering Academic & Student Affairs
Dr. Daniel McCarville, Professor of Practice, ASU School of Computing, Informatics, and Decision Systems Engineering
Dr. Ryan Meuth, Lecturer, ASU School of Computing, Informatics, and Decision Systems Engineering
Project Description
The NASA Psyche mission plans to launch a spacecraft to visit and conduct a scientific investigation of the asteroid (16) Psyche in our solar system’s asteroid belt. The primary motivator of this mission is to test the hypothesis that (16) Psyche is the core of a planetesimal (the precursor to a fully formed planet). To prepare for scientific investigations at Psyche, meteorite scientists wish to analyze the bulk mineral composition of iron meteorite samples, including those in the large collections housed at ASU and the Smithsonian. Standard analysis techniques are labor intensive, prompting the need for improved methods. The Iron Meteorite Imaging System (IMIS) was developed to capture high quality images of prepared meteorite samples for composition analysis using techniques from computer vision and machine learning. Image capture presented several challenges, including supporting a wide variety of meteorite sample sizes, handling glare and overexposure, and the consistency and accuracy of images. A prototype IMIS was developed, addressing these issues to allow semi-automated image capture.This sets the stage for future work, specifically on having the system extract information on bulk composition of samples from images.