Heuristic Assessment of Parameters of the Local Ground Approximation from Terrestrial LIDAR Data

摘要

The recently proposed quality measure is used to find the local ground approximation (LGA) from the data measured with the terrestrial laser scanning (TLS). The measure is the number of points in a thin layer located directly above the approximating surface. It can be optimized with the hill climbing algorithm. The method is robust against the data which can be treated as outliers from the ground model, so the TLS data have neither to be filtered nor segmented into those pertaining to the ground and to other objects (trees or lower vegetation). The results are compared with those obtained with the Hough transform-based method and assessed visually, with a positive result. If the ground does not depart too far from the assumed planar shape, the errors are small in relation to those obtained with other measurement modalities.