基于形态学算子和独立成分分析的高光谱遥感分类.pdf

Classification of Hyperspectral Data from Urban Areas using Morpholgical Preprocessing and Independent Component Analysis 1 1 1 2 Jon Aevar Palmason , Jon Atli Benediktsson , Johannes R. Sveinsson and Jocelyn Chanussot 1Department of Electrical and Computer Engineering University of Iceland, Hjardarhagi 2-6, 107 Reykjavik, Iceland e-mail {jaep, benedikt, sveinsso }@hi.is 2Signal Images Laboratory - LIS / INPG BP 46 - 38402 St Martin d’Heres - FRANCE e-mail jocelyn.chanussot@lis.inpg.fr Abstract— Classification of high-resolution hyperspectral data to principal component analysis but is more powerful and is investigated. Previously, in classification of high-resolution capable of finding the underlying factors or sources when the panchromatic data, simple morphological profiles have been con- principal component approach fails. ICA defines a generative structed with a repeated use of morphological opening and closing operators with a structuring element of increasing size, starting model for the observed multivariate data, which is typically with the original panchromatic image. This approach has recently given as a large database of samples. In the model, the data been extended for hyperspectral data. In the extension, principal variables are assumed to be linear mixtures of some unknown components of the hyperspectral imagery have been computed latent variables, and the mixing system is also unknown. in order to produce an extende