基于点云法矢变化的点云简化方法研究  

Data Point Reduction Research Based on Normal Vector Change

作　　者：张晶[1] 杨云生[1] 丰少伟[1] 

机构地区：[1]海军工程大学,武汉430033

出　　处：《计算机与数字工程》2011年第12期20-22,共3页Computer & Digital Engineering

基　　金：湖北省自然科学基金项目(编号:2009CDB087)资助

摘　　要：为了提高实体反求的效率,提出一种点云快速简化方法。该方法基于空间六面体栅格分割点云,依据栅格的拓扑结构搜寻测量点的邻域点,并采用一种三角面片的加权算法估算出每个测量点的法矢,并由此计算出每一个测量点到微切平面的距离,通过设置相应的法矢角度及距离阈值来对点云进行简化。实验表明该方法能够明显提高数据简化的效率。To improve the efficiency of entity reverse building, this paper proposed a method to reduce cloud data quickly. This method subdivided point cloud into many cubic grids, searched the nearest neighbors of a measured point based on topologized structure of cubic grids, and adopted a self-adjustable algorithm based on triangle surface to compute the normal vectors of every metrical point. Then the distance between the metrical point and local plane was computed, and the data point was reduced according to the set bonnet value of normal angle and distance. The result shows that this method can obviously impove the efficiency of data cloud reducing.

关 键 词：反求工程 点云 栅格划分 法矢 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]