Foundation laser radar data registration method based on single tree position and surface information

Abstract

The invention relates to an automatic registration method between multi-station foundation laser radar data based on single tree position and surface information and is mainly applied to the three-dimensional forest scene reconstruction based on multi-station foundation laser radar data. By the automatic registration method between multi-station foundation laser radar data, in the automatic alignment process, an additive foundation laser radar posture measuring device is used, the number of registration parameters required by registration need to be determined to reduce the number of control points required by registration. Simultaneously the similarity of a triangle formed by a plurality of single trees is utilized to recognize the homonymy single trees automatically, and control points are provided for the registration. By the automatic registration method between multi-station foundation laser radar data based on single tree position and surface information, the accuracy requirements and dependence for the control points in the registration process can be reduced, and the automatic registration of the multi-station foundation laser radar data under a forest environment can be realized.

Description

technical field [0001] The present invention relates to an automatic registration method between ground-based laser radar data based on single tree position and surface information. 3D forest scene reconstruction from lidar data. Background technique [0002] Ground-based lidar data have been gradually applied to detailed measurements of individual tree structural parameters, including the location and size of individual trees (Hopkinson et al. 2004; Henning and Radtke 2006). At the stand scale, the spatial distribution of forest canopy components is mainly considered (Henning and Radtke 2006). Due to occlusion between individual trees, it is difficult to accurately describe the forest using ground-based lidar data at the stand level using single-station data. Multi-site data needs to be used, which requires registration of multi-site data. In the process of archaeology and 3D reconstruction of buildings, since the research object has clear corner points and other feature...

AI Technical Summary

Problems solved by technology

Although this method is more suitable when the surface features are obvious, the stand density is low, and the understory vegetation is less; when the surface features are unknown, a single tree is surrounded by dead branches, and when the occlusion is serious, the registration required The control point is more difficult to choose, and the chance of success is greatly reduced

Images