Self-similarity Based Multi-layer DEM Image Up-Sampling

As one of the basic data of GIS, DEM data which expresses the surface elevation data is widely used in many fields. How to obtain a wide range of high-precision elevation data is a big challenge, the simple interpolation algorithm currently used is less accurate. Due to the fractal data characteristics of terrain data, DEM data shows strong self-similarity. Based on this feature, this paper proposes a multi-layer Dem image up-sampling method. Image up-sampling is performed multiple times in layers on the low-resolution DEM image, therefore, high-precision DEM information with less error is obtained. In this paper, elevation data of 30 m is expanded to elevation data of 10 m by gradually using this method. Experimental results show that the algorithm can achieve good results and has a small deviation from the real elevation data of 10 m.