Automatic Rock Detection and Classification in Natural Scenes

Autonomous geologic analysis of natural terrain is an important technique for science rovers exploring remote environments such as Mars. By automatically detecting and classifying rocks, rovers can efficiently survey an area, summarize results and concentrate on unusual discoveries, thus maximizing the scientific return. Similar techniques can be used in natural settings on land and under water where characterization of context is important to robot behavior. In this work, techniques for characterizing rocks using albedo, color, texture and shape properties are developed. These and other features are used to detect rocks, using a superpixel segmentation algorithm to locate and delineate the rocks with an accurate boundary. Once segmented, a Bayesian procedure is applied to geologically classify rocks. This capability is fundamental in enabling robots to conduct scientific investigations autonomously. Experiments demonstrate the usefulness of the albedo, color, texture and shape measures and establish the accuracy in computing two important geologic shape metrics, sphericity and roundness. The high performance of the rock detection and segmentation algorithm is analyzed and the results of geologic rock classification are presented. These methods successfully perform geologic rock analysis, while giving an indication of what features are most important for this and similar tasks. As color, texture and shape are such powerful features, this approach is also applicable to detecting, segmenting and classifying other natural objects and terrain features. Thesis supervisor: Prof. David Wettergreen