A New Approach to Urban Modelling Based on LIDAR

Estimating building forms from LIDAR data has usually been d o e by attempting to fit standardized building types to the residual data points after an estimated bare earth ter ain surface has been removed. We propose an approach based on segmenting the raw data into high and low regions, an d the modelling the walls and roofs by extruding the triangulated terrain surface (TIN) using CAD-type Eule r operators. The segmentation may be done by the addition of building boundary data to the TIN so as to force tr iangle edges to match the boundaries, and then using Euler operators to extrude the building, producing ve rtical walls rather than the more usual sloping walls formed from TIN models alone. If boundary data is not availab le then an automated segmentation method based on adaptive Voronoi cells may be used, so that each cell conta ins either high or low LIDAR data, but not both. This prismatic model, with flat-topped cells, approximates the b uilding forms without hypothesizing specific building types. Once the segmentation is achieved, and the walls cons tructed, we attempt to model the roofs by calculating the eigenvalues and eigenvectors of the vector normals of th e TIN sections within the building boundaries. The smallest eigenvalue gives a predicted roof orientation, an d the resulting roof profile is then modelled.