Quality Inspection And Accuracy Evaluation For DEM

Spatial data quality has a direct impact on the application of GIS. DEM is one of Nation's Fundamental Geographic Information, and its quality has been increasingly becoming people's focus.For any DEM project, precision, efficiency, and benefit are three main factors to be considered. Among these factors, precision is most important. Provided that precision can't be guaranteed, efficiency and benefit can't be gained. Therefore, the precision of DEM has gained more and more researcher's attention. The research of DEM's quality has both theory and application significance.Based on the systematic analysis of error's generation, propagation and uncertainty, and detailed summarization and analysis of many DEM quality inspection and precision valuation models, this thesis mainly researched into gross error detection for grid DEM. Its main contents are summed up as follows.1) The thesis had brought robust estimation into bend surface fitting algorithm and compared the effect of least squares surface fitting algorithm with robust estimation of least squares surface fitting algorithm.2) The thesis also put forward a method of gross error detection considering the topographic characteristic, the method extracted topographic feature point and employed subsection curve fitting algorithm and robust estimation of least squares surface fitting algorithm for topographic feature point and common topographic point respectively.It is shown that robust estimation of least squares surface fitting algorithm for grid DEM can reduce gross errors to a certain extent. So it has a better result too. It has reduced the influence of topographic feature point on DEM gross error detection to employ different gross error detection methods for topographic feature point and common topographic point respectively.Furthermore, we also developed an application system of DEM quality inspection and precision evaluation with Visual Basic6.0+ArcEngine on the basis of topographic maps. Besides basic GIS functions, the system has realized some special functions which are listed as follows: gross errors detection for grid points, interpolation precision evaluation, chart junction precision evaluation and contour superposition precision evaluation, as well as creating hypsometric maps, hill-shading map, 3D DEM and other visualized assistant inspection functions.Among which, we have done some deep research into gross errors detection for grid points, interpolation precision evaluation, contour superposition precision evaluation. And the results show that, for contour superposition precision evaluation, the superposition of original contour and derived contour has performed much better and had a more completed and reliable result of DEM precision evaluation, and the result is only and stable too.