摘要
在三维激光点云数据配准的过程中,利用传统Iterative Closest Point(ICP)算法搜索对应点对时速度慢,而且配准精细化程度低,远达不到三维建模后期处理的要求。针对这一问题,提出一种基于KDTree改进的ICP算法以实现激光点云数据的快速精细化配准。通过实验验证算法的有效性和合理性,为后期模型重建过程中的三角网格化、曲面化、纹理映射提供强有力的理论和实践基础。
In the process of three-dimensional laser point Cloud data registration, the time of using the traditional ICP algorithm to search the corresponding point is slow, and the precision degree of registration is low, as far as less than the requirements of three-dimensional reconstruction＇ post-processing. In order to solve this problem, it comes up with an improved ICP algorithm based on KDTree for rapid refinement registration laser point Cloud data. Through the experimental verification, the algorithm is effective and reasonable, and it provides a strong theoretical and practical foundation for the triangular mesh, curved surface, texture mapping of the process of model reconstruction.
出处
《微型机与应用》
2015年第14期81-83,86,共4页
Microcomputer & Its Applications
