Finding Images of Terrain Features in Large Databases Collected by Rovers - Robotics Institute Carnegie Mellon University

Finding Images of Terrain Features in Large Databases Collected by Rovers

Debra Schreckenghost, Tod Milam, David Lees, and Terrence W. Fong
Conference Paper, Proceedings of International Symposium on Artificial Intelligence, Robotics and Automation in Space (iSAIRAS '12), September, 2012

Abstract

NASA’s exploration and scientific missions will produce terabytes of information. A key challenge for these missions is improving utilization of this information. For example, geologists are often interested in finding images of terrain features in an area. The DELVE (Database Exploration to Locate Visual Elements) software searches an image database to find the images most likely to show a terrain feature. It does this by determining whether the feature location is within the expected viewing area of the camera when the image was taken. In this paper we describe the DELVE software and the evaluation of DELVE using images collected at a NASA field test in 2010.

BibTeX

@conference{Schreckenghost-2012-7586,
author = {Debra Schreckenghost and Tod Milam and David Lees and Terrence W. Fong},
title = {Finding Images of Terrain Features in Large Databases Collected by Rovers},
booktitle = {Proceedings of International Symposium on Artificial Intelligence, Robotics and Automation in Space (iSAIRAS '12)},
year = {2012},
month = {September},
keywords = {planetary exploration, field robots, planetary rover},
}