Genetic feature selection combined with composite fuzzy nearest neighbor classifiers for hyperspectral satellite imagery

For high-dimensional data, the appropriate selection of features has a sign ificant effect on the cost and accuracy of an automated classifier. In this paper, a feature selection technique using genetic algorithms is applied. For classification, crisp and fuzzy k-nearest neighbor (kNN) classifiers ar e compared. Composite fuzzy classifier architectures are investigated. Expe riments are conducted on airborne visible/infrared imaging spectrometer (AV IRIS) data, and the results are evaluated in the paper. (C) 2002 Elsevier S cience B.V, All rights reserved.