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：2000
卷号：XXXIII Part B7(/1-4)
页码：1389-1394
出版社：Copernicus Publications
摘要：This paper focus on the use of spectral sensors data in conjuction with GIS databases. The paper describes the cooperative use of spatial information and remote sensing imagery to improve image analysis as well as fusion of cues from each da- tasource for land-use classification purposes. As the analysis is performed within a digital environment, a raster approach is proposed. In order to integrate knowledge in the digital environment, a declarative form was selected to represent in- formation. A rule based system, in which knowledge is represented using conditional expressions and syntax, composed of antecedents, logical combination of propositions and consequences, containing a collection of decisions, is here used. Since the envolved concepts are vague, the use of less formal logic to represent knowledge is best suited to handle them. Therefore, the adopted strategy is based on fuzzy logic combination rules. First, cues about land-use classes are obtained from each datasource. Since the nature uf both datasets are different, a common base is established using fuzzy representation of information. In this manner, the information derived from each datasource, GIS and satellite imagery, can be combined using fuzzy logic
