期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2005
卷号：XXXVI-8/W27
出版社：Copernicus Publications
摘要：Building and tall object shadowing results in a relevant number of pixels in HR hyperspectral urban data to be insufficiently illuminated to get a reliable estimate of the local reflectance and thus to be correctly classified. Masking away shadowed pixels is one of the simplest approaches to this issue. In this paper we instead attempt to investigate the features of these pixels in order to exploit their possible information content. Tests are conducted on an hyperspectral dataset acquired in July 2002 by the DLR over the town of Pavia using the DAIS and ROSIS sensors. We show how a careful selection of the classes may attain significant increases in the classification accuracy. Although the data set is limited and thus the results not to be made general, we are satisfied that those pixel have an exploitable information content
关键词：high resolution; hyperspectral data; paving materials; shadow pixels; urban areas
