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/W52
页码：326-331
出版社：Copernicus Publications
摘要：Due to the possibility of acquiring precise height data of large areas rapidly, airborne laser scanning systems are particularly suitable for obtaining information about the damage situation immediately after a disaster in large scale. This paper presents a technique for the detection and classification of damages occurring on buildings in affected areas. It is based on the comparison of pre-event building models composed of planar surfaces with planar surfaces extracted from laser scanning data acquired directly after the disaster. In a first step, segments are created by superposing the pre- and post-event surfaces. Subsequently, for every segment the geometrical characteristics of the corresponding pre- and post-event surface are compared. Finally, the segments are assigned to damage types using a fuzzy logic classification approach. The results achieved for each processing step by applying the method on data containing real building damages are presented and analysed
