A subway leakage water detection method based on point cloud information

Abstract

The application discloses a subway leakage water detection method based on point cloud information, utilizes three-dimensional laser point cloud data to segment a tunnel, segments tunnel ring numbers into multiple single-ring point cloud models, uniformly changes point cloud scalar values in each single ring, corrects the range of gray values, replaces the gray values with point cloud intensity values, sets a threshold value, counts all ring piece leakage water areas and proportions, screens ring pieces with larger water seepage areas, and achieves the effect of rapidly detecting a leakage water area. The application directly utilizes intensity values of point cloud information to distinguish the leakage water area. An image is composed of pixel points, generally is converted into an image through gridding of point cloud data, different materials have different light reflection intensities, the intensity values can be directly utilized as gray values for leakage water distinguishing, thus the calculation process can be simplified, the recognition speed is improved, and the application has strong popularization value.

Description

Technical Field

[0001] This invention belongs to the field of tunnel engineering technology, and specifically designs a method for detecting water leakage in subway tunnels. Background Technology

[0002] The main construction method for subway tunnels is the shield tunneling method, and the main problem that endangers the health of shield tunnels is water leakage.

[0003] Water leakage in subway tunnels can lead to weathering, spalling, and corrosion of the lining, damaging the lining structure. Leakage can also soften and deform the surrounding rock. Since leakage often occurs at anchor bolt holes, it corrodes the bolts, accelerating equipment aging. In harsh winters, water seepage in tunnel cracks can freeze, causing the cracks to expand and widen. Icicles formed from the leakage can also intrude into the tunnel's clearances, affecting train safety. If water leakage is not detected and repaired promptly, its erosion of the lining will become increasingly severe. Over time, this can lead to displacement or even misalignment of the tunnel lining segments, and in severe cases, segment spalling or even tunnel collapse.

[0004] Currently, tunnel leakage detection mainly relies on manual inspections, which may lead to missed detections within limited inspection windows. To address this issue, Chinese invention patent application number 201810066937.7 discloses a method for automatically identifying leakage in tunnel images obtained from 3D scanner point cloud information and processed images. Chinese invention patent number 202010967797.8 similarly uses convolutional neural networks from deep learning algorithms for image recognition of leakage defects, employing automated and batch processing for defect detection. Chinese invention patent number 202210071582.7 is also based on deep learning algorithms, using a VGG network to train the dataset and a CNN network to identify defect locations. Existing methods primarily locate leakage locations through image recognition, requiring photographs of leakage defects as input samples. This process involves long fieldwork hours and requires certain lighting conditions, increasing the workload. Furthermore, deep learning algorithms suffer from low accuracy in training data, affecting the final generalization ability.

[0005] Traditional methods for detecting water leakage in subway tunnels often rely on manual inspections and photographic data collection, consuming significant time for data processing and report compilation. These traditional methods fail to meet the requirements of efficiency, accuracy, and safety, necessitating new technologies to improve the speed and automation of tunnel water leakage detection. 3D laser scanners, capable of acquiring point cloud location and intensity information simultaneously, are utilized to propose a simple, efficient, and automated method for detecting water leakage in subway tunnels. Summary of the Invention

[0006] To address the aforementioned issues, this invention provides a method for detecting water leakage in subway tunnels based on point cloud data, thereby resolving the problems of slow detection speed and low automation in existing technologies for detecting water leakage in subway tunnels.

[0007] The technical solution of this invention is a method for detecting water leakage in tunnels, particularly for shield tunnels, comprising the following steps:

[0008] (1) Obtain subway tunnel point cloud information using a 3D laser scanner, including the point cloud's position (x, y, z) and scalar value (Sclar). The x-axis is perpendicular to the tunnel's direction, the y-axis is the tunnel's forward direction, the z-axis is vertical, and Sclar is a scalar information inherent to the scanner, reflecting the reflection intensity value of the point cloud information.

[0009] (2) Divide the point cloud data along the tunnel axis according to the width of the tunnel segment, and divide a tunnel point cloud model into a single-ring point cloud model according to the ring number.

[0010] (3) Correct the intensity of point cloud information on each ring piece, and correct the intensity value to 0-255 and replace the gray value with the intensity value;

[0011] (4) By taking advantage of the difference in strength between water and the tunnel surface, a threshold is set to filter out all point clouds within the ring that are less than the threshold, which are identified as seepage point clouds, and the percentage of seepage area of ​​a single ring is calculated.

[0012] (5) The surface area of ​​a single ring can be calculated if the number and type of single ring segments are fixed. The leakage area of ​​a single ring can be calculated by combining the proportion of leakage area. The leakage area of ​​all rings within a certain distance can be counted according to the ring number to quickly detect defects.

[0013] This invention utilizes 3D laser point cloud data to segment tunnels, dividing the tunnel ring number into multiple single-ring point cloud models. By uniformly changing the scalar values ​​of the point clouds within each single ring to correct the range of grayscale values, and replacing the grayscale values ​​with point cloud intensity values, a threshold is set. The leakage area and percentage of all ring segments are statistically analyzed, and ring segments with larger leakage areas are selected, achieving the effect of rapid detection of leakage areas.

[0014] One innovation of this invention is that, in the process of detecting and treating water leakage in shield tunnels, the intensity values ​​of point cloud information are directly used to distinguish leakage areas. Images are composed of pixels, and point cloud data is typically converted into images through rasterization. Different materials reflect light differently, and the intensity values ​​can be directly used as grayscale values ​​to distinguish leakage areas, thus simplifying the calculation process and improving recognition speed. Simultaneously, determining the location of the leakage area based on point cloud location information improves the efficiency and accuracy of real-time processing.

[0015] This invention applies three-dimensional laser scanning technology to the detection of water leakage in subway tunnels, enabling real-time monitoring of tunnel cross-sectional data, greatly improving detection efficiency and accuracy, and bringing convenience to subway construction, design, and construction parties. It has strong promotional value. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of the subway water leakage detection method based on point cloud data provided by the present invention;

[0018] Figure 2 This is a schematic diagram illustrating the cross-sectional cutting principle according to a specific embodiment of the present invention;

[0019] Figure 3 This is a schematic diagram of a subway tunnel divided into a single ring according to a specific embodiment of the present invention;

[0020] Figure 4 This is a statistical chart of the leakage area and percentage according to an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art are within the scope of protection of the present invention.

[0022] According to an embodiment of the present invention, a method for detecting water leakage in subway tunnels is provided.

[0023] like Figure 1 As shown, the method for detecting water leakage in subway tunnels according to an embodiment of the present invention includes the following steps:

[0024] 101. Obtain point cloud data of subway tunnels using a 3D laser scanner.

[0025] 102. Preprocess the raw point cloud data.

[0026] 103. The tunnel ring point cloud model is generated by segmenting according to the ring number.

[0027] 104, Correction of point cloud intensity within the ring image.

[0028] 105. Set a threshold to filter out areas with leakage and calculate the percentage of leakage area.

[0029] 106. Traverse all the rings into which the detection area is divided, and count the leakage area of ​​each ring. Complete the tunnel leakage identification.

[0030] The present invention, through the above technical solution, first acquires subway tunnel point cloud data using a 3D laser scanner, then preprocesses the raw point cloud data, then segments the tunnel point cloud model according to ring number, then corrects the point cloud intensity within each ring, sets a threshold to calculate the proportion and area of ​​water leakage in a single ring, and finally traverses the detection section to count the water leakage area according to ring number to complete the water leakage identification.

[0031] This invention can accurately identify the location, area, and quantity of water leakage in subway tunnels, greatly improving identification accuracy and efficiency. Furthermore, the three-dimensional laser scanning technology uses a non-contact measurement method and requires no calibration, is less affected by the environment, and can collect data at all times, even in the absence of light, thus improving detection efficiency.

[0032] Point cloud data preprocessing includes point cloud denoising, point cloud stitching, point cloud simplification, and coordinate transformation to generate new point cloud data.

[0033] In one embodiment, the point cloud model acquired by the laser scanner is first preprocessed, and the processed point cloud model is then segmented according to ring numbers.

[0034] Furthermore, based on the center coordinates of the starting and ending rings of the detection section, a normal vector perpendicular to each ring along the tunnel axis is established. Then, a cross-sectional equation is established based on the center coordinates of each ring. Finally, using the point-to-surface distance formula, the point cloud data is segmented sequentially according to the ring number, as follows: Figure 2 As shown.

[0035] The specific calculation process is as follows: The equation of the cross-section passing through the center of the ring segment is established based on the center coordinates of the initial ring number.

[0036] (x s -x e (xx) n )+(y s -y e (yy) n )+(z s -z e (zz) n ) = 0

[0037] Where (x) s ,y s ,z s (x) represents the coordinates of the starting ring center. e ,ye ,z e ) Termination ring center coordinates, (x n ,y n ,z n The coordinates of the center of the nth ring piece.

[0038] Given a segment width d (twice the distance from the point to the surface), the set of points Φ(x) belonging to that segment is... i ,y i ,z i The following equation must be satisfied:

[0039]

[0040] The tunnel point cloud model for the detection section is divided into n single rings according to the above formula, such as... Figure 3 As shown.

[0041] Furthermore, the point cloud intensity within the ring is corrected by converting the point cloud intensity value to a grayscale value of 0-255.

[0042] The specific calculation process is as follows: first, the point cloud intensity values ​​within the ring are corrected.

[0043]

[0044] in, D represents the spatial distance of the point cloud within the ring. s The reference distance is I′ (usually the tunnel radius), and the corrected intensity value is I′, while the original point cloud intensity value is I′.

[0045] Furthermore, the corrected intensity values ​​are normalized to 0-255 as the grayscale values ​​of the point cloud within the ring.

[0046]

[0047] Where H is the grayscale value of the point cloud, and I′ max I′ represents the maximum value of the point cloud intensity after correction within the ring. min The minimum point cloud intensity after correction within the ring.

[0048] Furthermore, based on research data on water leakage image recognition, a water leakage threshold K is set (generally taken as (50-70) after binarization). Areas with gray values ​​less than the threshold K are considered water leakage areas. The proportion ρ of the water leakage area within a single ring is calculated.

[0049] In one example, for a single-ring point cloud model generated from the tunnel point cloud model of the detection section, the single-ring leakage area is calculated, and the leakage area is statistically analyzed according to the ring number. Specifically, this includes the following steps:

[0050] Based on the segment width, the surface area S of the segment within a single ring can be determined. By filtering the point cloud intensity values, the proportion of the leakage area of ​​a single ring segment can be calculated, thereby calculating the leakage area of ​​that ring segment.

[0051] Furthermore, the leakage surface area of ​​the ten ring plates within the detection section can be calculated, and a point-line diagram of ring number and leakage area can be established, as shown below. Figure 4 As shown.

[0052] Furthermore, based on the correspondence between ring number and mileage, a dot-line graph of mileage-leakage area can be established. By setting the maximum leakage area, mileage sections with severe leakage problems can be automatically screened out.

[0053] The method for detecting subway water leakage of the present invention generates a grayscale image of the tunnel surface lining by acquiring point cloud data. Taking the grayscale image as the research object, it skips digital image processing technology and directly processes the tunnel point cloud information to complete the identification of water leakage and count its location, number, area and other information, thereby realizing the automated detection of subway water leakage.

[0054] In summary, by utilizing the technical solution of this invention, point cloud data of the tunnel is acquired, preprocessed, and then segmented according to the tunnel segment thickness to generate a single-ring point cloud model. The point cloud intensity within each ring is corrected, and intensity is used as the grayscale value. A grayscale threshold is set to calculate the leakage area of ​​each ring. The leakage area is statistically analyzed according to the ring number, and the location, area, and number of leaks are obtained based on the relationship between the ring number and the mileage. The advantage of introducing a 3D laser scanner is that it is not limited by the environment or light source, reducing the need for personnel while improving the efficiency and automation of the inspection.

[0055] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for detecting water leakage in subways based on point cloud information, applicable to shield tunnels, characterized in that: Includes the following steps, (1) Obtain subway tunnel point cloud information through a 3D laser scanner, including the position (x, y, z) and scalar value of the point cloud; where the x-axis is the direction perpendicular to the tunnel direction, the y-axis is the tunnel forward direction, the z-axis is the vertical direction, and Sclar is a scalar information provided by the scanner, which reflects the reflection intensity value of the point cloud information; (2) Divide the point cloud data along the tunnel axis according to the width of the tunnel segment, and divide a tunnel point cloud model into a single-ring point cloud model according to the ring number; (3) Correct the intensity of point cloud information on each ring piece, and correct the intensity value to 0-255 and replace the gray value with the intensity value; (4) By taking advantage of the difference in strength between water and the tunnel surface, a threshold is set to filter out all point clouds within the ring that are less than the threshold, which are identified as seepage point clouds, and the percentage of seepage area of ​​a single ring is calculated. (5) The surface area of ​​a single ring can be calculated if the number and type of single ring segments are fixed. Combined with the proportion of leakage area, the leakage area of ​​a single ring can be calculated. Based on the ring number, the leakage area of ​​all ring numbers within a certain distance can be counted to quickly check for defects. First, the point cloud model acquired by the laser scanner is preprocessed, and the processed point cloud model is segmented according to the ring number. Then, based on the cross-sectional center coordinates of the starting and ending rings of the detection section, a normal vector perpendicular to each ring along the tunnel axis is established. Based on the center coordinates of each ring, a cross-sectional equation is established, and the point cloud data is segmented according to the ring number using the point-to-surface distance formula. The specific calculation process is as follows: The equation of the cross-section passing through the center of the ring segment is established based on the center coordinates of the initial ring number. ； in To determine the starting coordinates of the ring center, The coordinates of the center of the termination ring are as follows: The coordinates of the center of the nth ring piece; Given a segment width d, the set of points belonging to that segment is... Satisfy the following formula: ； The point cloud model of the detection section tunnel is divided into n single rings according to the above formula.

2. The subway water leakage detection method based on point cloud information according to claim 1, characterized in that: The tunnel was segmented using 3D laser point cloud data. The tunnel ring number was divided into multiple single-ring point cloud models. The scalar values ​​of the point cloud in each single ring were uniformly changed to correct the range of gray values. The gray values ​​were replaced by point cloud intensity values. A threshold was set and the leakage area and proportion of all ring segments were statistically analyzed.

3. The subway water leakage detection method based on point cloud information according to claim 1, characterized in that: First, point cloud data of the subway tunnel is acquired using a 3D laser scanner. Then, the raw point cloud data is preprocessed, and the tunnel point cloud model is segmented according to the ring number. Next, the point cloud intensity in each ring is corrected, and a threshold is set to calculate the proportion and area of ​​leakage in a single ring. Finally, the detection section is traversed, and the leakage area is counted according to the ring number to complete the leakage identification.

4. The subway water leakage detection method based on point cloud information according to claim 1, characterized in that: Point cloud data preprocessing includes point cloud denoising, point cloud stitching, point cloud simplification, and coordinate transformation to generate new point cloud data.

5. The subway water leakage detection method based on point cloud information according to claim 1, characterized in that: The point cloud intensity within the ring is corrected, and the point cloud intensity value is converted to a grayscale value of 0-255; The specific calculation process is as follows: first, the point cloud intensity values ​​within the ring are corrected. ； in, D represents the spatial distance of the point cloud within the ring. s For reference distance, This is the corrected strength value. This is the point cloud intensity value before correction.

6. The subway water leakage detection method based on point cloud information according to claim 1, characterized in that: The corrected intensity values ​​are normalized to 0-255 and used as the grayscale values ​​of the point cloud within the ring. ； Where H is the grayscale value of the point cloud. This represents the maximum value of the point cloud intensity after correction within the ring. The minimum point cloud intensity after correction within the ring.

7. The subway water leakage detection method based on point cloud information according to claim 1, characterized in that: Based on research data on water leakage image recognition, a water leakage threshold K is set. Areas with gray values ​​less than the threshold K are considered to be water leakage areas. The proportion ρ of the water leakage area within a single ring is calculated.

8. The subway water leakage detection method based on point cloud information according to claim 1, characterized in that: For the single-ring point cloud model generated from the tunnel point cloud model of the detection section, calculate the single-ring leakage area and statistically analyze the leakage area according to the ring number; specifically, this includes the following steps: The surface area S of the inner segment of a single ring is determined based on the segment width. By filtering the point cloud intensity values, the proportion of the leakage area of ​​the single ring segment is calculated, thereby calculating the leakage area of ​​the segment in that ring.

9. A subway water leakage detection method based on point cloud information according to claim 1, characterized in that: A point-line graph of mileage and leakage area is created based on the ring number and mileage. By setting the maximum leakage area, the mileage section with serious leakage problems is automatically screened. The method for detecting subway water leakage generates a grayscale image of the tunnel surface lining by acquiring point cloud data. Taking the grayscale image as the research object, it skips digital image processing technology and directly processes the tunnel point cloud information to complete the identification of water leakage and count its location, number, and area information, thereby realizing the automated detection of subway water leakage.

Images