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：43-50
出版社：Copernicus Publications
摘要：This paper presents a new methodology for the automatic adaptation of image analysis object models for the extraction of 2D landscape objects to a lower image resolution. The knowledge of the object models is represented in form of semantic nets. The developed adaptation method includes a prediction of the object's behaviour in linear scale-space using analysis-by-synthesis. The scale behaviour prediction also takes into account scale events possibly occurring during scale change. The presented algorithm extends previous work concerning the adaptation of parallel linear object parts to object models consisting of 2D area object parts with arbitrary orientation. An example for the adaptation of an object model describing a 4-arm road junction with symbol markings demonstrates the application of the methodology. Finally, conclusions point out enhancements of the method
