Building Detection Using LIDAR Data and Multispectral Images

A method the automatic detection of buildings from LIDAR data and multispectral images is presented. A classification technique using various cues derived from these data is applied in a hierarchic way to overcome the problems encountered in areas of heterogeneous appearance of buildings. Both first and last pulse data and the normalised difference vegetation index are used in that process. We describe the algorithms involved, giving examples for a test site in Fairfield (Victoria).