摘要
针对当前三维图像重构效率差,重构结果遍历覆盖度低等问题,提出基于大数据可视化激光测距城市空间三维图像重构方法。利用激光测距仪采集大量城市空间三维点云元数据,对元数据实施序列化处理获取城市空间相关大数据。基于获取的数据进行可视化分析,点云数据中包含激光采样点数值范围,可构建城市空间数字表面模型;利用该模型获取城市地面高度均值,通过曲面拟合获取城市空间数字地面模型;根据上述两个模型划分城市空间地面建筑物,获取城市空间内建筑物三维信息。通过三角形网格法将三维点信息转换为三维面信息,实现城市空间三维图像重构。测试结果显示,所提方法三维图像重构加速比达到5.53,输出图像遍历覆盖最高可达98.83%,说明该方法可实现最初研究目的。
Aiming at the problems of poor efficiency of 3 D image reconstruction and low traversal coverage of reconstruction results, a 3 D image reconstruction method of urban space based on big data visualization laser ranging is proposed. A large number of 3 D point cloud meta data of urban space are collected by laser rangefinder, and the metadata is serialized to obtain big data related to urban space. Based on obtained data for visual analysis, the point cloud data contains the range of laser sampling points, which can be used to construct urban spatial digital surface model. Based on the model, the average urban ground height is obtained, and the urban spatial digital terrain model is obtained by surface fitting. According to the above two models, the urban space ground buildings are divided, and three-dimensional buildings information of urban space are obtained. The three-dimensional point information is transformed into three-dimensional plane information by triangle mesh method to realize its three-dimensional image reconstruction. The test results show that the speedup ratio of 3 D image reconstruction reaches 5.53, and the maximum traversal coverage of output image can reach 98.83%, which indicates that this method can achieve original research purpose.
作者
江平
JIANG Ping(Hubei University of Arts and Sciences,Xiangyang Hubei 441000,China)
出处
《激光杂志》
CAS
北大核心
2022年第3期174-178,共5页
Laser Journal
基金
湖北省自然科学基金(No.2017CFB591)。
关键词
大数据
可视化
激光测距
城市空间
三维图像
图像重构
big data
visualization
laser ranging
urban space
3D image
image reconstruction
