文章基本信息

标题：ON THE USE OF THE GEOSTATISTICAL ESTIMATION TECHNIQUES TO THE GENERATION, DEBUGGING AND ANALYSIS OF DIGITAL SURFACE MODELS
本地全文：下载
作者：Jorge Delgado
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 B4 (/1-3)
页码：206-213
出版社：Copernicus Publications
摘要：To utilise DSMs effectively it is necessary to ensure that they are sufficiently accurate to meet the requirements of our specific application. In order to assess the accuracy of a model two basic aspects must be considered: the measured data accuracy and the modelling quality. In this work we show a geostatistical approach to the surface modelling. The geostatistical methods are well suited to assessing accuracy and they allow treat the entire modelization process. These methods allow the study of phenomena that fluctuate in space. The process begins with a data capture using a dense grid in a representative zone of the study area. Those data are used for the spatial variability estimation, obtaining information about the influence zones and presence of anisotropies. The modeled spatial variability functions will be used for the final grid design considering the maximum admissible estimation error, that depends only of the spatial variability functions and the data positions –not data values-. Once the grid to be measured is defined, the final capture is made and the data are debugged using the cross validation method. Using the debugged data and the information derived from the structural analysis the modelization is made. The estimation will be optimum (non-biased and minimum estimation error). The estimation error will be known for the entire model and it will be a very important information for the evaluation of the model quality. Finally, we present an example of analysis using the Factorial Kriging method that allows the filtering using a structural space depending filter schema
关键词：Digital Surface Models; Data Processing; Mathematical Models; Modelling; Spatial Data