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：2008
卷号：XXXVII Part B4
页码：305-312
出版社：Copernicus Publications
摘要：Today, with the situation of rapidly emerging high resolution earth observation data by optical and microwave sensors there is a growing need for efficient methods to derive, maintain and revise land cover data at various scales by regional, national and European authorities. This paper focuses on parts of a current research project named DeCOVER, namely the change detection which is used to identify candidates of change in land-use (LU) or land-cover (LC). The change information needed is automatically derived from multi-temporal satellite image data of a spatial resolution of approximate 5m to be comparable to the planned German RapidEye system. This paper concentrates on the first outcomes of the implemented process chain, namely the focusing module and the segmentation and classification module. The focusing module has two tasks: First, to detect potential changes relevant to the GIS database and second, to decide, whether the detected changes and the affected GIS-objects subsequently can be reclassified automatically in the segmentation and classification module or must be processed manually. Different change indicators are implemented based on a comparison of the input satellite data of two different points of time. These indicators in combination with a transition-probability-matrix are used to steer the subsequent process of verifying the indicated changes according to the sorted probabilities of change. As a result a change-layer is produced which outlines potential changes relevant to the GIS database (here the nomenclature of DeCOVER) and which holds evidence about the plausibility of each detected change
