Mars Image Content Classification: Three Years of NASA Deployment and Recent Advances
This work provides a practical solution for NASA PDS users to efficiently find specific Mars images within a vast dataset, addressing a data accessibility problem for researchers and the public.
This paper describes the development and deployment of content-based classification and search capabilities for millions of Mars orbital and surface images hosted by the NASA Planetary Data System. These systems, publicly accessible via the PDS Image Atlas, have been in deployment for three years.
The NASA Planetary Data System hosts millions of images acquired from the planet Mars. To help users quickly find images of interest, we have developed and deployed content-based classification and search capabilities for Mars orbital and surface images. The deployed systems are publicly accessible using the PDS Image Atlas. We describe the process of training, evaluating, calibrating, and deploying updates to two CNN classifiers for images collected by Mars missions. We also report on three years of deployment including usage statistics, lessons learned, and plans for the future.