Enter Note Done
Go to previous page in this tab
Session
  • Presentation | P31C: Machine Learning and Data Science Methods for Planetary Science II Oral
  • Oral
  • Share Icon
    Map Icon
    Bookmark Icon
  • P31C-03: Grain-Scale Morphometrics of Martian Sedimentary Deposits via Neural Network Segmentation of Rover Imagery
    • Schedule
    Notes
  • Location Icon294
    NOLA CC
    Set Timezone
    •  
    View Map

Generic 'disconnected' Message
Author(s):
Ayden Hayes, Dartmouth College (First Author, Presenting Author)
Ari Koeppel, Dartmouth College
Emma Rogers, Dartmouth College
Mansa Krishna, Dartmouth College
Shreya Gandhi, Dartmouth College


Mars rovers have taken thousands of close-up images of the planet’s surface, revealing tiny grains of sediment. These grains hold clues about how they were moved and what Mars’s environment was like in the past. But there are millions of grains, and measuring them by hand would be prohibitively time-consuming and prone to human error.


To make this process faster and more consistent, we used computer algorithms that can learn to recognize patterns in images in order to find and measure grains in rover images. By matching grain characteristics such as its size and shape, with known types of landscapes—like riverbeds, lakes, wind-blown areas, and volcanic zones—we can learn more about where these grains came from and how they got there. This helps us better understand Mars’ surface history and past climate. The tool we built is free and open to the public to support future Mars research.




Scientific Discipline
Suggested Itineraries
Neighborhood
Type
Main Session
Discussion