A marked point process for automated building detection from lidar point-clouds

Abstract

This letter presents a novel algorithm for automated building detection from light detection and ranging (lidar) point-clouds. The algorithm takes advantage of a marked point process to model the locations of buildings and their geometries. A Bayesian paradigm is used to obtain a posterior distribution for the marked point process. A Reversible Jump Markov Chain Monte Carlo (RJMCMC) algorithm is implemented for simulating the posterior distribution. Finally, the maximum a posteriori (MAP) scheme is used to obtain an optimal building detection. The results obtained on a set of lidar point-clouds demonstrate the efficiency of the proposed algorithm in automated detection of buildings in complex residential areas.

Acknowledgements

The authors would like to acknowledge the provision of the Downtown Toronto data set by Optech Inc., First Base Solutions Inc., GeoICT Lab at York University, and ISPRS WG III/4. The Vaihingen data set was provided by the German Society for Photogrammetry, Remote Sensing, and Geoinformation (DGPF).