摘要
激光扫描建模中通常采用一次匹配技术,但是室外三维景观处于三维环境下,一次匹配会导致大量激光点云数据特征丢失,导致室外景观三维建模效果差。提出基于二次匹配技术的室外景观三维激光建模方法。首先,在三维激光扫描技术中引入SFM方法,以此获取完整的点云数据,选取控制点进行初始配准,通过ICP算法精确配准两种点云数据。为提高点云数据的融合性,针对数据颜色等特征进行二次匹配,实现点云数据融合。最后,使所有点云数据在同一坐标中,将融合过的点云数据进行去噪、滤波和点云拼接等预处理,将数据中的两种曲面进行拟合并导入三维软件中,实现室外景观建模。试验结果表明,所提方法的三维景观建模立体水平较高,建模效果清晰,精度较高。
One-time matching technology is usually used in laser scanning modeling. However, when outdoor 3 D landscape is in 3 D environment, one-time matching will lead to the loss of a large number of laser point cloud data features, resulting in poor 3 D modeling effect of outdoor landscape. A three-dimensional laser modeling method of outdoor landscape based on secondary matching technology is proposed. Firstly, SFM method is introduced into 3 D laser scanning technology to obtain complete point cloud data. Control points are selected for initial registration, and two kinds of point cloud data are accurately registered by ICP algorithm. In order to improve the fusion of point cloud data, secondary matching is carried out for data color and other characteristics, and point cloud data fusion is realized. Finally, all the point cloud data are in the same coordinate, the fused point cloud data is preprocessed by denoising, filtering and point cloud splicing, and the two kinds of surfaces in the data are fitted and imported into 3 D software to realize outdoor landscape modeling. Experimental results show that the three-dimensional landscape modeling with the proposed method has high stereo level, clear modeling effect and high precision.
作者
高连平
宋培娟
王庄
Gao Lianping;Song Peijuan;Wang Zhuang(Changchun Guanghua University,Changchun,Jilin 130033,China;Jilin Normal University,Siping,Jilin 136000,China)
出处
《应用激光》
CSCD
北大核心
2022年第3期147-153,共7页
Applied Laser
基金
2020年吉林省教育厅人文社科研究项目(JJKH20211421SK)
激光与物质相互作用国家重点实验室基金(SKKLIM1612)。
关键词
三维扫描
激光点云
二次匹配
数据融合
三维建模
曲面拟合
three-dimensional scanning
laser point cloud
secondary matching
data fusion
3D modeling
surface fitting
