基于图像分割的地铁隧道渗漏水快速检测

Rapid detection of subway tunnel leakage based on image segmentation

地铁隧道具有线路长、地质环境复杂、环片数量多、检测作业时间短等特点。隧道结构渗漏水病害检测是保障地铁隧道安全的重要工作,传统人工巡查检测方法检测速度慢,需耗费大量的人力物力,不能满足地铁更好运营需求。为提高隧道渗漏水检测效率,更好地适应轨道交通运营需求。本文依托高速移动三维激光扫描技术获取点云数据,提出局部图像分割方法,获取隧道渗漏水病害信息。首先通过空间变换法将三维点云数据转换为二维正射影像,采用图像二值化处理算法增强隧道渗漏水病害区域的边缘信息;然后利用区域描述算法对图像内的渗漏水区域进行分析,获取渗漏水病害区域的大小和里程信息,从而达到自动化识别隧道结构病害。研究结果表明,本文方法可快速检测地铁隧道内渗漏水病害的位置、面积等信息,节约检测时间和检测成本的同时保证了检测结果的准确性。

Subway tunnels are characterized by long lines,complex geological environment,large number of rings,and short inspection operation time.The detection of water leakage in the tunnel structure is an important task for subway tunnel safety.The traditional manual inspection method has a slow detection speed and requires a lot of manpower and material resources,and cannot meet the better needs of subway operations.In order to improve the efficiency of tunnel leakage detection and better adap to the needs of rail transit operations.This paper relies on high-speed moving three-dimensional laser scanning to obtain point cloud data,and proposes a partial image segmentation method to obtain tunnel water leakage disease information.Firstly,the three-dimensional point cloud data is converted into two-dimensional orthoimages through the spatial transformation method,and the image binary processing algorithm is used to enhance the edge information of the tunnel leakage disease area;Then the area description algorithm is used to analyze the water leakage area in the image,and obtain the size and mileage information of the water leakage disease area,so as to achieve automatic identification of the tunnel structure disease.The research results show that the method in this paper can quickly detect the location and area of water leakage diseases in subway tunnels,and save the detection time and cost.At the same time ensure the accuracy of the detection results.