激光雷达点云特征分析与数据分割

发布时间：2019-06-16 12:42

【摘要】：激光雷达扫描技术是一种采集三维数据、重建三维模型的新兴手段，随着激光雷达技术的发展，和三维模型重建在各个领域的需求，点云数据的处理已经是国内外研究的热点问题。论文结合固定式的地面激光雷达扫描系统以及其采集的数据作为研究对象，分别深入研究激光雷达点云特征和点云数据的分割算法。本文首先研究了激光雷达扫描系统的原理，介绍激光雷达系统的组成，通过对激光雷达采集传回的数据进行处理获得的点云数据，设计软件完成点云的显示，保存数据作为后续数据处理分析的基础。其次按照点云数据处理的一般流程完成对点云数据预处理工作，得到处理方便的数据集合。然后，根据点云几何特征估计法线和曲率，同时提出点云特征直方图的描述子，根据法线和曲率特征对点云数据提取特征点。最后，总结点云分割的理论，分析并实现点云数据分割的算法，比较不同算法在点云数据上分割作用的效果。
[Abstract]:Lidar scanning technology is a new means to collect 3D data and reconstruct 3D model. With the development of lidar technology and the demand of 3D model reconstruction in various fields, the processing of point cloud data has become a hot issue at home and abroad. Combined with the fixed ground lidar scanning system and the collected data as the research object, the characteristics of laser radar point cloud and the segmentation algorithm of point cloud data are deeply studied in this paper. In this paper, the principle of lidar scanning system is studied, and the composition of lidar system is introduced. by processing the point cloud data collected and returned by lidar, the software is designed to display the point cloud, and the data is saved as the basis of subsequent data processing and analysis. Secondly, according to the general process of point cloud data processing, the point cloud data preprocessing is completed, and a convenient data set is obtained. Then, the normal and curvature are estimated according to the geometric features of the point cloud, and the descriptors of the point cloud feature histogram are proposed, and the feature points are extracted from the point cloud data according to the normal and curvature features. Finally, the theory of point cloud segmentation is summarized, the algorithm of point cloud data segmentation is analyzed and implemented, and the effect of different algorithms on point cloud data segmentation is compared.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN957.52

【参考文献】