移动激光扫描技术下的隧道横断面提取及变形分析

Tunnel Cross Section Extraction and Deformation Analysis Based on Mobile Laser Scanning Technology

移动激光扫描技术具有获取数据速度快、点密度大、精度高、无需拼站等优点,本文利用该技术获取的点云数据对隧道形变检测方法进行了研究和应用。首先,结合边界检测算法和随机抽样一致性算法计算隧道的空间中轴线,在此过程中省去了传统方法中对大量点云进行旋转的过程。其次,采用k近邻法计算各断面的缓冲区对点云进行分块,并利用投影法构建断面点集,提高了断面点集获取的速度。最后,提出了一种迭代椭圆拟合去噪的方法,在拟合出断面线的同时去除了噪声。最终通过与全站仪获取的断面数据进行对比验证了该方法的精度并将其应用在某段圆形盾构隧道上,采用隧道检测小车获取点云数据,每隔2 m截取一个断面,然后通过与理论断面的对比分析了隧道的形变情况。

Mobile laser scanning technology has the advantage of fast data acquisition, high density and high precision, in addition, it doesn＇t need to stitch the adjacent station after scanning. Based on mobile laser scanning data, the method of tunnel deformation detection and analysis is studied. Firstly, the spatial central axis of the tunnel is calculated by combining the boundary detection algorithm with random sampling consistency algorithm, and the rotation process of a large number of point clouds in traditional method is eliminated. Secondly, the k-nearest neighbor method is used to compute the buffer of each section to block the point clouds, and the sectional point sets are automatically constructed by projection method, increased the calculation speed of cross sections. Finally, an iterative elliptic fitting method is proposed to denoise the cross sections and fit the sectional curve lines.The precision of the method is verified by comparing With cross section obtained by total station.The proposed method is applied on a section of circular shield tunnel, which is intercepted into cross sections in the interval of 2 m based on the point clouds obtained by tunnel scan, and deformation analysis is performed by comparing the cross sections with theoretical one.