Recently vehicle-borne laser scanning system equipped with linear array laser scanner, GPS and INS has played an
important role on the acquisition of urban spatial data. In this paper, a method was proposed for the segmentation of
building facades from vehicle-borne laser scanning data in the outdoor urban environment, which contains five steps
including 1) preliminary filtering with depth histogram analysis; 2) binarization of point cloud; 3) mathematical
morphology processing of binary image; 4) segmentation of binary image and 5) extracting building façades with the
mapping relationships of vehicle-borne laser scanning data and binary image. Depth histogram analysis, calculation of
grid size and determination of the number of dilation were described in detail. Finally, an experimental study was made
to demonstrate the feasibility of the proposed method for the segmentation of urban building facades.
In this paper we focus on feature extraction method of complex objects from range data captured by vehicle-borne
laser scanning system. In our study, we classified the objects obtained by laser scanner into three types: road surface,
building facade and free sharps which consisted by scattered points. We have developed the corresponding methods for
the extraction of those objects. The DSCD (Density Statistics with Conformal Division) for road surface extraction and
DSEI (Density Statistics with Equal Interval) for building facade extraction. Especially, our experiment has shown that
the MST (Minimum Spanning Tree) is a good way for scattered points classification, and the free sharps are extracted
with high automation and efficiency.