A Color Interest Operator for Landmark-Based Navigation

Abstract

Landmark-based approaches to robot navigation require an "interest operator" to estimate the utility of a particular image region as an effective representative for a scene. This paper presents a color interest operator consisting of a weighted combination of heuristic scores. The operator selects those image regions (landmarks) likely to be found again, even under a different viewing geometry and/or different illumination conditions. These salient regions yield a robust representation for recognition of a scene. Experiments showing the reproduceability of the regions selected by this operator demonstrate its use as a hedge against environmental uncertainties. Introduction One important ability of natural visual systems is that they spend most of their time on "interesting" portions of their input, that is, on those aspects of an image which inform the task at hand. The Stanford Cart had one of the first artificial vision systems which looked for regions of interest from the scenes it...