Complete classification of raw LiDAR data and 3D reconstruction of buildings

LIDAR (LIght Detection And Ranging) data are a primary data source for DTM (Digital Terrain Model) generation and 3D city models. This paper presents a three-stage framework for a robust automatic classification of raw LIDAR data as buildings, ground and vegetation, followed by a reconstruction of 3D models of the buildings. In the first stage the raw data are filtered and interpolated over a grid. In the second stage, a region growing technique is used to segment the data in sets with a step edge along the region border. In the third stage a rule-based scheme is applied for the classification of the regions, using their geometric and topological relationships. Finally, polyhedral building models are reconstructed analysing the topology of building outlines, building roof slopes and eaves lines. Results obtained on datasets with different ground point density, gathered over the town of Pavia (Italy) with Toposys and Optech airborne laser scanning systems are shown to illustrate the effectiveness of the proposed approach.