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：2006
卷号：XXXVI Part 3
出版社：Copernicus Publications
摘要：Today's processes to extract man-made objects from measurement data are quite traditional. Often, they are still point based, with the exception of a few systems which allow to automatically fit simple primitives to measurement data. At the same time, demands on the data are steadily growing. The need to be able to automatically transform object representations, for example, in order to generalize their geometry, enforces a structurally rich object description. Likewise, the trend towards more and more detailed representations requires to exploit structurally repetitive and symmetric patterns present in man-made objects, in order to make extraction cost-effective. In this paper, we address the extraction of building fac .ades in terms of a structural description. As has been described previously by other authors, we use a formal grammar to derive a structural fac .ade description in the form of a derivation tree. We introduce two new concepts. First, we use a process based on reversible jump Markov Chain Monte Carlo (rjMCMC) to guide the application of derivation steps during the construction of the tree. Second, we attach variables and constraint equations to the symbols of the grammar, so that the derivation tree automatically leads to a constraint equation system. This equation system can then be used to optimally fit the entire fac.ade description to given measurement data
关键词：Markov Chain; constraint equations; fac ; .ade modelling; building extraction; least squares adjustment
