数字高程模型误差及其评价的问题综述  

A Review on the Issues in DEM Error and DEM Quality Assessment

作　　者：吴艳兰[1] 胡海[1] 胡鹏[1] 庞小平[1] 

机构地区：[1]武汉大学资源与环境科学学院,武汉市珞喻路129号430079

出　　处：《武汉大学学报（信息科学版）》2011年第5期568-574,共7页Geomatics and Information Science of Wuhan University

基　　金：国家自然科学基金资助项目(40701155);国家863计划资助项目(2009AA12Z224)

摘　　要：从数字高程模型(DEM)传递误差、基于中误差的DEM误差模型及其主要问题、DEM误差分布实验和DEM内插误差新认识几个方面分析了当前DEM误差研究的主要进展,用"中误差"讨论DEM传递误差是建立在测量误差传递理论基础之上的,但沿用"中误差"来讨论DEM内插模型逼近误差和DEM整体误差却缺乏理论依据。DEM误差分布的空间相关性实验对DEM中误差评价法所应具备的随机误差性提出质疑,却可以用基于逼近理论的DEM内插模型来解释,说明用逼近误差理论研究DEM内插误差的途径是正确、可行的。We provide an overall review of the present standing of research in digital elevation model（DEM） error,focusing mainly on the four aspects： DEM propagation error,DEM error models based on the measure of mean square error（MSE） and the problems that have arisen in the using of MSE measure,DEM error distribution experiments,and new understanding of DEM interpolation error.It addresses the following viewpoints： the foundation of the use of MSE measure to assess DEM propagation error is error propagation theory,however,using MSE measure to assess DEM interpolation error and the total error lacks a theoretical basis.On the other hand,the spatial patterns of error distribution shown in many DEM error experiments have cast considerable doubt on the usefulness of MSE-based measures,but the spatial property of DEM error can be explained by the DEM error model based on approximation theory.It is correct and feasible to apply approximation theory to discuss DEM interpolation error.

关 键 词：DEM误差 传递误差 函数逼近误差 DEM质量控制 

分 类 号：P208[天文地球—地图制图学与地理信息工程]