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：83-86
出版社：Copernicus Publications
摘要：In the study of GIS/LIS, two kinds of data structureare studied: raster data structure and vector data structure.It is proved by experience and theory that the vectordata has a higher compression rate and a well-organizedlogical structure ( that is, an object is represented by itsboundary polygons. , and a polygon is a series of vectorsVI, V2, ..., VN , where Vi and Vi+! have a same endpoint), and some basic image operations , such as scaling,rotating, calculating the area and perimeter , can beeasily done on the vector data structure. But the spatialorganization of the vector data is very loose , which leadsto low efficiency when querying the objects near or at agiven position. So, a good spatial index is needed to speedup the response of the system.In this paper, it is suggested that the vector data can beorganized by a quadtree -like index structure - - - -QO (quadtree of objects). The definition of QO ; the algorithmfor creating QO of a image, the algorithms of operationon QO are given. By analysing the querying effectivenessof QO, it is shown that the time consume ofqueries on the vector with QO - index is much moresmaller than that on the "pure" vector data
