Study On Theory And Method For Quality Control Of DEM Spatial Data

Digital Elevation Model(DEM) is a method of digital description and simulation for earth surface. It is a important part of basic establishments of spatial data. The application of DEM is more and more abroad. It can be used not only in survey analysis but also modeling and analysis of diversified no elevation attribute data that distributes in series and changed gradually on two dimension geography space. DEM is a disperse expression of series terrain surface that is based on sampling data. So original data's precision determines DEM quality directly. If spatial data have error, the height interpolated by these data will be likely to be distorted seriously . Any interpolation method can all not remedy original dada's infection to DEM quality. So it is the research of DEM quality assess and control that has very important significance.Based on expatiating on DEM construction theory by the numbers and contrasting analysis the whole bag of tricks in DEM spatial data interpolation, on this paper, DEM error source will be analyzed. And in virtue of error spread principle, the error spread of linear interpolation, cubic spline interpolation and double orientation interpolation will be deduced and analyzed. These methods provide gist for DEM quality assess. In allusion to this state that sampling data are liable to be polluted by abnormity error, combining equivalence weight principle, robust inverse distance weight, robust moving approach robust universal kriging have be brought forward. These robust interpolation methods will be used to resist abnormity error's infection. They are the foundation for quality control of DEM spatial data.