摘要
三维点云数据配准拼接和坐标转换是高精度重构建筑物的重中之重,深入研究三维点云数据的建筑物重构方法,采用三维激光扫描点云数据配准与坐标系转换方法,完成不同视角下建筑数据的多视定位,并把数据转换至同一坐标系下后,对建筑物三维点云数据进行预处理,获取高分辨率建筑三维点云数据;匹配和重构建筑物高分辨率三维基础点云数据,完成建筑物重构。结果表明:在重构平顶建筑时,所提方法平均偏差比是1.4%,重构尖顶建筑时,最大重构误差是5%,说明所提方法通用性较好,重构精度较高;且重构最大分辨率为1080 PPI,可实现三维点云数据的建筑物重构。
3D point cloud data registration,splicing and coordinate conversion are the top priority of high-precision reconstruction of buildings. After in-depth study on the building reconstruction method of 3D point cloud data,the 3D laser scanning point cloud data registration and coordinate system conversion methods are adopted to complete the multi view positioning of building excitation scanning data from different perspectives,and the data is converted to the same coordinate system,Preprocess the building 3D point cloud data to obtain high-resolution building 3D point cloud data;matching and reconstructing high-resolution 3D basic point cloud data of buildings to complete building reconstruction. The experimental results show that the average deviation ratio of the proposed method is 1. 4% when reconstructing flat roof buildings and the maximum reconstruction error is 5% when reconstructing pointed roof buildings,indicating that the proposed method has good universality and high reconstruction accuracy;moreover maximum resolution of reconstruction is1080 PPI,which can realize the building reconstruction of 3D point cloud data.
作者
王群
WANG Qun(Xi'an Siyuan University,Xi'an 710038,China)
出处
《粉煤灰综合利用》
CAS
2022年第6期123-128,共6页
Fly Ash Comprehensive Utilization
基金
陕西省科技厅项目(2020JQ-917)。
关键词
高分辨率
三维点云数据
建筑物重构
最大重构误差
high resolution
3D point cloud tracing data
building reconstruction
maximum reconstruction error
