Automated Registration of Unorganised Point Clouds from Terrestrial Laser Scanners

Abstract

Terrestrial laser scanners provide a three-dimensional sampled representation of the surfaces of objects resulting in a very large number of points. The spatial resolution of the data is much higher than that of conventional surveying methods. Since laser scanners have a limited field of view, in order to obtain a complete representation of an object it is necessary to collect data from several different locations that must be transformed into a common coordinate system. Existing registration methods, such as the Iterative Closest Point (ICP) or Chen and Medioni's method, work well if good a priori alignment is provided. However, in the case of the registration of partially overlapping and unorganised point clouds without good initial alignment, these methods are not appropriate since it become very difficult to find correspondence. A method based on geometric primitives and neighbourhood search is proposed. The change of geometric curvature and approximate normal vector of the surface formed by a point and its neighbourhood are used to determine the possible correspondence of point clouds. Our method is tested with a simulated point cloud with various levels of noise and two real point clouds.