Orientation of Airborne Laser Scanning Point Clouds with Multi-View, Multi-Scale Image Blocks

摘要: Comprehensive 3D modeling of our environment requires integration terrestrial and airborne data, which is collected, preferably, using laser scanning photogrammetric methods. However, these multi-source data accurate relative orientations. In this article, two methods for solving orientation problems are presented. The first method includes registration by minimizing the distances between an point cloud a model. model was derived from measurements points. used as reference validation. Having completed in object space, images known. second utilizes interactive multi-scale image block cloud. both aerial images. Experiments with revealed that accuracy increased when more were included block. orientations compared. comparison showed correct rotations most difficult to detect accurately method. Because forces fit images, inaccurate cause corresponding shifts positions. test case, differences only shifts, could solve repeatedly within couple centimeters.