Scaleless Conversion Study Of DEM Based On Wavelet

Scale has been the focus of geographical scientific problems.With the continuous expansion and increasing levels of demand in GIS applications, the use of multi-scale implementation of spatial data analysis of multi-level observation of the geographical situation is becoming increasingly important. To achieve multi-level observation, analysis of the geographical situation making use of multi-scale spatial data is becoming increasingly important. Multi-scale spatial data processing and expression has become the core content of the setting up of spatial database. The multi-resolution analysis of wavelet is extremely similar to the connotation of multi-scale spatial data.Therefore this paper has researched DEM-scale conversion using wavelet algorithm.This paper has accomplished four conversion in scale and derived a series of different level of detail of the DEM through adding the details of different wavelet coefficient - 0,25%, 50%, 75% by using cofi4 system wavelet with the sample of the Loess Plateau-Chunhua and Jiuyouangou District 1: 10000DEM. The accuracy of derived DEM is evaluated and the information of the terrain -the slope, roughness, terrain curvature, gully density, water, etc is extracted from the derived DEM analyzing of the impact on topographical factors when added different details information. Finally, the changes law of landforms information is reached and the changes information is quantitative assessed, when added different details of wavelet coefficient. The study found that Selecting a different of details of information is not only reservation or highlight the details of topography, but also making the auto-general of landscape continuous transform in the upper and lower levels and achieving scaleless DEM.Finally, the stream network has been extracted before and after the General DEM and the fractal model of the expansion of the root has been applied to the quantitative integrated model of river network to evaluate the level of detail derived DEM of different scale by model of square root law.