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：2007
卷号：XXXVI-3/W49A
页码：123-128
出版社：Copernicus Publications
摘要：In mountainous areas, the landcover extraction out of orthoimages through semi-automatic classification is limited by several factors (such as large shadow areas, radiometric similarities between different themes, inhomogeneous radiometry among regions of the same class...). Image information is not sufficient to separate the different classes. Nevertheless, good results can be obtained by dividing each landcover class "c" into two subclasses "c in shadow" / "c not in shadow" and introducing external information in the classification process. This information can be an older or more generalized database, geographic knowledge concerning the links between relief and landcover, or prior information concerning shadows. This external knowledge is then interpreted in terms of a priori probabilities and merged with radiometric information from the image in a MAP per region classification process. Besides, the results can also be improved by the use of combinations of channels calculated from the initial image
关键词：Image classification - Landcover extraction - Data fusion - Uncertainty management - National base map - Landcover- ; knowledge based interpretation - Mountainous landcover extraction - MAP classification method
