Dense DTM Generalization Aided by Roads Extracted from LiDAR Data

Abstract

The paper concerns the generalization of DTM extracted from LiDAR data. The essence of generalization is reducing details while enhancing important features at the same time; so for the purpose of terrain surface visualization special attention has to be given to the enhancement and generalization of topographic objects like dams, roads etc. The focus of this work is laid on the extraction of road objects and their contribution into the enhancement of the generalized terrain model. An algorithm for the extraction of roads is developed and is followed by a generalization algorithm that weights together road networks and filtered LiDAR point clouds. Following the presentation of the algorithm results for this approach are shown and evaluated.