Abstract

This paper presents the Plannett system, which combines artificial neural networks to achieve expertlevel accuracy on the difficult scientific task of recognizing volcanos in radar images of the surface of the planet Venus. Plannett uses ANNs that vary along two dimensions: the set of input features used to train and the number of hidden units. The ANNs are combined simply by averaging their output activations. When Plannett is used as the classification module of a three-stage image analysis system called JARtool, the end-to-end accuracy (sensitivity and specificity) is as good as that of a human planetary geologist on a four-image test suite. JARtool-Plannett also achieves the best algorithmic accuracy on these images to date. Introduction Between 1991 and 1994, the Magellan space probe collected more than 30,000 synthetic aperture radar (SAR) images of the surface of the planet Venus, a greater amount of data than all previous planetary missions combined (Smyth et al. 1995). To a...

Citations

2304	Learning Internal Representations by Error Propagation - Rumelhart, Hinton, et al. - 1986
354	Neural network ensembles, cross validation, and active learning - Krogh, Vedelsby - 1995
141	Combining forecasts: A review and annotated bibliography - Clemen - 1989
114	An adaptive clustering algorithm for image segmentation - Pappas - 1992
56	Combining Estimators Using Non-Constant Weighting Functions ", to appear - Tresp, Taniguchi - 1995
35	Automating the hunt for volcanoes on venus - Burl, Fayyad, et al. - 1994
29	Inferring ground truth from subjective labelling of Venus images - Smyth, Fayyad, et al. - 1994
20	Learning probabilistic relational concept descriptions - Ali - 1996
14	Image Database Exploration: Progress and Challenges - Fayyad, Smyth - 1993