Mars Image Content Classification: Three Years of NASA Deployment and Recent Advances

TL;DR

This paper discusses the development, deployment, and evaluation of CNN-based classifiers for Mars images over three years, enhancing search capabilities in NASA's PDS Image Atlas.

Contribution

It introduces CNN classifiers tailored for Mars images, detailing their training, deployment, and real-world usage over an extended period.

Findings

CNN classifiers improved image search efficiency

Three years of deployment provided valuable usage insights

Lessons learned inform future Mars image classification efforts

Abstract

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.