摘要
基于点云分割和区域生长的点云法向矢量提取系统,应用Li DAR数据的回波强度标定点云分割结果内的目标强度,采用区域增长的思想实现目标点云法向矢量的提取,未配准点云边界数据,提取效率差。设计大数据下激光点云边界法向矢量精确提取系统,利用激光点云图像采集模块获取激光点云图像,将点云图像存储至自动存储模块中,通过应用模块依据点云图像数据采用软件中的算法提取云边界法向矢量;系统软件采用大数据框架下激光点云边界数据提取的云模型分类和配准激光点云边界数据,采用Sobel方向掩盖法划分点云边界数据,提取激光点云边界上随意一点的法向矢量。经实验证明,相较于其他系统,所设计系统法向矢量生成和提取的平均时间分别是18 s和11 s,平均提取精度约为97. 94%,说明该系统具有较高的提取精度和提取效率。
Based on the point cloud segmentation and regional growth of the point cloud normal vector extraction system,the echo intensity of lidar data is used to target the intensity of the target cloud segmentation results,and the idea of regional growth is used to achieve the target point cloud normal vector extraction. The extraction efficiency is poor due to the lack of accurate cloud boundary data. A normal vector extraction system for laser point cloud boundary under big data is designed,and laser point cloud image acquisition module is used to obtain the point cloud image, and then be stored in an automatic storage module. The normal vector of the boundary is extracted by the algorithm in software based on the point cloud image data. The system software uses the cloud model classification and registration of the laser point cloud boundary data extracted by the laser point cloud boundary data under the framework of big data. The Sobel direction mask method is used to divide the point cloud boundary data,and extracts the normal vector at a random point on the laser point cloud boundary. Experiments show that compared with other systems,the average time for generating and extracting normal vectors is 18 s and 11s,respectively,and the average extraction accuracy is about 97. 94%,indicating that the system has higher extraction accuracy and efficiency.
作者
陈晓燕
CHEN Xiaoyan(Neijiang Normal University,Sichuan Neijiang 641100,China)
出处
《激光杂志》
北大核心
2019年第10期65-69,共5页
Laser Journal
基金
四川省科技厅应用基础研究计划(No.2015JY0119)
关键词
大数据
激光点云边界
法向矢量
提取
分类
配准
big data
laser point cloud boundary
normal vector
extraction
classification
registration
