Two algorithms for extracting building models from raw laser altimetry data

Two new techniques for the determination of building models from laser altimetry data are presented. Both techniques work on the original laser scanner data points without the requirement of an interpolation to a regular grid. Available ground plan information may be used, but is not required. Closed solutions for the determination of the parameters of a standard gable roof type building model based on invariant moments of 2 1/2-D point clouds are shown. In addition, the analysis of deviations between point cloud and model does allow for modelling asymmetries such as dorms on a gable roof. By intersecting planar faces nonparametric buildings with more complex roof types can also be modelled. The techniques were applied to a FLI-MAP laser scanner dataset covering an area of 500 x 250 m(2) with a density of more than 5 points/m(2). Within this region, all but one building could be modelled. An analysis of the variance of the parameters within a group of buildings indicates a precision in the range of 0.1-0.2 m. (C) 1999 Elsevier Science B.V. All rights reserved.