This doctoral thesis addresses the assessment of the geometric similarity of point clouds, which are registered with the use of artificial targets. We can usually accurately define the coordinates from the point clouds obtained from airborne laser scanning of artificial targets or by a classic measurement within a chosen coordinate system. The main goal of this thesis is to research the use of cone shaped artificial targets in the assessment process of point clouds. By performing two practical tests we have managed to determine the best geometry of the cone for use in laser scanning procedures. The use of cones as ALS targets is improved with the automatic point recognition procedure with the RANSAC method. We need to define the parameters and additional constants for the physical models of the cones used in ALS procedures. From the analysis of the point clouds obtained by two test airborne laser scans we defined the quality of the reflections given by the cones and defined the criteria for the qualitative use of such targets. We described and presented the results of the established referential geodetic network in Radovljica, which we used in order to evaluate the quality of the point clouds obtained from an airborne laser scan, i.e. their geometric similarities. Through various analyses we verified the accuracy and precision of the created models and the coordinates of the control points on the various reflective surfaces. The usefulness of cones as artificial targets is defined by comparing the quality of the point clouds on control points obtained from various reflective surfaces when registered with cones, black and white targets and other ground points.
|
