M-Split Estimation in Laser Scanning Data Modeling

This paper proposes a new approach to 3D modeling on the basis of laser scanner data. It presents the attempt to automatically detect two planes and an edge in one processing algorithm. The algorithm is based on the M _split estimation, which is a recently developed regression method of multi functional models. Plane equations in three dimensional space are derived on the basis of the presented method. M _split estimation divides the functional model of least squares adjustment in particular into two groups. Both simulated and real data from airborne laser scanning is tested to present the results. In each case the mehod gave good results which encourages for further work on the application of this method for 3D modeling.