Lidar Point Clous Data Processing And3D Terrain Resgistration Method Research

3D terrain registration process has great significance for natural disaster monitoring and evaluation, uav autonomous navigation and other related fields. The results of3d terrain registration accuracy and its processing speed are greatly affected by the original terrain data. In order to improve the registration results of precision and the speed of data processing, this paper puts forward a method that extracts3D terrain from Lidar point clouds data and then carry out3D terrain registration process.The lidar data after pretreatment, filtering, interpolation can generate3D terrain data (DEM) with high precision.The coordinate system of DEM datas has more or less deviation because the DEM datas have different time phase, different angles.They must be unified into a coordinate system before the application analysis of DEM datas.And then we can complete3d terrain registration through a series of algorithms.In this paper, main work is as follows:1) The Lidar data filtering algorithm of domestic and foreign existing was studied, the various classic filtering algorithms are implemented in the MFC framework, and it compares their advantages and disadvantages;2) Studying the domestic and foreign classical interpolation algorithm,it compares the advantages and disadvantages of various kinds of interpolation algorithms, and have implemented them in the software;3) Contrast analysis of the3D terrain registration algorithms at home and abroad, the advantages and disadvantages of those algorithms, and compares the adaptability of each algorithm, and proposes a kind3D algorithms without control points based on Sift algorithm;4) Studiying the3D terrain data storage formats at home and abroad, and completing the transformation of those common data formats with each other. It enhances the adaptability and robustness of the3D terrain matching algorithm which is proposed in this paper.