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:1992
卷号:XXIX Part B3
页码:727-734
出版社:Copernicus Publications
摘要:Computer interpretation of images requires a decisionon what to look for and a strategy exploiting thisknowledge. A generic model for which objects make upthe image is designed, assum ing low altitude aerialimages. The generic model is exploited first to choose arepresentation of the image suitable for parsing,secondly for defining a parsing procedure. Asrepresentation, a segmentation is chosen consisting ofsegment boundaries plus descriptions of segmentinteriors. Segment boundaries are only of the kindrecognized by the parsers, in this case straight line andsmooth curve segments represented using strip trees.Secondly, this segmentation is used as an input to a setof parsers which use a list of properties of buildings inorder to interpret the input. Both line parsers and regionparsers are used. Each parser is successful for a limitedtask. A set of parsers is scanned until the parse isaccepted in a consistency test. This strategy is chosenas it is considered easier to test if a parse is acceptablethan to design a well performing general parser.
关键词:Image interpretation; image segmentation;parsing; knowledge based interpretation