A comparative study of DEM registration approaches

Abstract

The registration process forms an important step in the integration of Digital Elevation Models (DEMs) of differing accuracy and resolution. Registration effectively constitutes a surface-to-surface matching in which horizontal and vertical offsets between the two DEMs are determined and corrected. This paper evaluates five approaches for DEM registration, each being classified according to the sequence of estimating offsets. Furthermore, an efficient search strategy is presented to assure a rapid and robust registration. The approaches were tested using four datasets of differing resolution and accuracy, each being registered to a LiDAR DEM covering two test sites of varying topography and land cover. The accuracy assessment results show that the use of GMI for horizontal registration followed by weighted averaging for vertical registration of DEMs is an approach that can be recommended.