Abstract: The safety and longevity of subway operations are significantly impacted by water leaks.However,traditional manual inspection methods are inadequate for inspecting busy subway lines.Therefore,a fast and accurate inspection technology is needed.In this study,a LiDAR-based leak detection technique is proposed.The method divides the point cloud data along the tunnel axis direction based on the tube sheet width.Additionally,a section of the tunnel point cloud model is divided into multiple single-ring point cloud models according to the ring number.To eliminate the influence of laser distance and laser incident angle on the identification results,intensity correction is applied.The point cloud data of each ring is processed by the image processing algorithm,which enables automatic identification and statistical analysis of water leakage diseases based on the tunnel mileage calculated from the ring number information.On the basis of this methodology,a software for identifying leakage water was developed using the Python programming language.The software was verified using engineering cases.The results show that the method presented in this paper can quickly and accurately identify the location and size of water leakage.It has significant practical applications.