基于三维点云的隧道全局中线提取方法及应用

发布时间：2018-02-24 16:21

本文关键词： 隧道工程中线提取三维点云随机采样一致性算法　出处：《湖南大学学报(自然科学版)》2017年09期 　论文类型：期刊论文

【摘要】：提取隧道横断面通常是基于隧道中线,而隧道中线提取是最困难的一个环节.本文提出了一种有效的隧道全局中线自动提取算法:通过从隧道的点云水平投影中提取边界点,并利用随机采样一致性算法提取边界点中用于多模型拟合的内点,最终的隧道中线基于再次随机采样的全局优化算法确定.所提出的算法被应用于位于四川的写字岩隧道,结果表明:全局中线提取达到了较高的精度,其均方根误差为16.5mm;与全站仪测量的隧道中线点相比,偏差的均方根误差为10.3mm.
[Abstract]:The extraction of tunnel cross section is usually based on the tunnel center line, and the tunnel center line extraction is one of the most difficult links. In this paper, an effective algorithm for automatic tunnel global midline extraction is proposed: the boundary point is extracted from the horizontal projection of the tunnel point cloud. The random sampling consistency algorithm is used to extract the interior points from the boundary points for multi-model fitting, and the final tunnel midline is determined based on the global optimization algorithm of re-random sampling. The proposed algorithm is applied to the writing rock tunnel in Sichuan Province. The results show that the global center line extraction has a high accuracy, the root mean square error is 16.5 mm, and the deviation root mean square error is 10.3 mm compared with the tunnel midline point measured by the total station.
【作者单位】：西南交通大学交通隧道工程教育部重点实验室;西南交通大学土木工程学院;
【基金】：国家自然科学基金重点项目(U1434206)~~
【分类号】：TP391.41;U452.1

【相似文献】