A method and system for detecting misalignment between rings of a shield tunnel

Abstract

This invention provides a method and system for detecting inter-ring misalignment in shield tunnels. The method includes: transforming three-dimensional point cloud data of the tunnel into two-dimensional planar point cloud data; converting the two-dimensional planar point cloud data into a grayscale image of the tunnel; extracting annular joint images of the tunnel based on the brightness value of each pixel in the grayscale image; denoising the annular joint images to obtain denoised annular joint images; and obtaining the inter-ring misalignment amount of the tunnel based on the denoised annular joint images. This invention, by converting the three-dimensional point cloud data of the tunnel into a grayscale image and then obtaining the inter-ring misalignment amount based on this image, can quickly and comprehensively collect data from inside the tunnel, greatly improving the detection efficiency of inter-ring misalignment.

Description

Technical Field

[0001] This invention belongs to the field of tunnel engineering measurement technology, and more specifically, it relates to a method and system for detecting inter-ring misalignment in shield tunnels. Background Technology

[0002] In recent years, to address traffic congestion caused by urbanization, major cities across China have accelerated the planning and construction of rapid, high-capacity transportation infrastructure, primarily subways and light rail. As of January 2022, the total mileage of urban rail transit in China reached approximately 8,708 km. Subway projects are characterized by long construction cycles, large investments, and are considered long-term, heritage-building projects, requiring stringent quality standards at every stage. Accidents can cause severe social impacts and significant economic losses. However, due to technical and construction conditions, as well as a lack of timely maintenance, subway tunnels almost universally suffer from structural defects of varying degrees during operation, with tunnel deformation and misalignment being among the most common.

[0003] Current methods for detecting tunnel structural cross-sections primarily involve using total stations or cross-sectional instruments to measure each section and measuring point individually. The results are then compared with design values ​​or previous measurements to verify the tunnel construction quality and deformation. However, these existing methods only reflect localized characteristics, making it difficult to efficiently conduct large-scale inspections. Furthermore, they are highly dependent on the skill level of the personnel and environmental conditions, making it challenging to guarantee inspection quality. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for detecting inter-ring misalignment in shield tunnels, aiming to solve the problem of low detection efficiency in existing detection methods.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A method for detecting inter-ring misalignment in a shield tunnel includes the following steps:

[0007] Step 1: Obtain the 3D point cloud data of the tunnel;

[0008] Step 2: Project the three-dimensional point cloud data onto a standard cylinder, and obtain two-dimensional planar point cloud data according to the unfolded shape of the standard cylinder;

[0009] Step 3: Convert the two-dimensional planar point cloud data into a tunnel grayscale image;

[0010] Step 4: Extract the tunnel annular joint image based on the brightness value of each pixel in the tunnel grayscale image;

[0011] Step 5: Denoise the tunnel annular joint image to obtain a denoised tunnel annular joint image;

[0012] Step 6: Obtain the inter-ring misalignment of the tunnel based on the denoised tunnel annular joint image.

[0013] Preferably, step 2: projecting the three-dimensional point cloud data onto a standard cylinder, and obtaining two-dimensional planar point cloud data according to the unfolded shape of the standard cylinder, includes:

[0014] Centered on the center point of the fitted ellipse, the point cloud data is projected onto the standard design cross-section of the tunnel to obtain the point cloud data projection points; wherein, the cross-section of the standard cylinder is the standard design cross-section of the tunnel; the point cloud data projection points are:

[0015]

[0016]

[0017] Where P' = (x', z'), P' is the projection point of the point cloud data, R is the tunnel radius, P(x, z) is any point cloud on the tunnel, and O(x... o , z o () represents the center point of the fitted ellipse. It is the positive unit vector along the Z-axis;

[0018] Two-dimensional planar point cloud data is obtained from the projection points of the point cloud data, including:

[0019] Formula used:

[0020]

[0021]

[0022] Two-dimensional planar point cloud data is obtained; where ∠P'OZ is and The included angle, P”=(x”,z”) is the point after P' is expanded, x” is the x-coordinate of point P” on the two-dimensional plane point cloud, z” is the y-coordinate of point P” on the two-dimensional plane point cloud, R is the tunnel radius, x’ is the x-axis coordinate of the projection point P’, x o To fit the x-coordinate of the center point of the ellipse, z o The ordinate of the center point of the fitted ellipse is given.

[0023] Preferably, step 3: converting the two-dimensional planar point cloud data into a tunnel grayscale image includes:

[0024] Step 3.1: Divide the two-dimensional planar point cloud data into grids of a preset length;

[0025] Step 3.2: Use the mean value of the intensity information of all point clouds in each grid as the pixel value of the corresponding grid to obtain the tunnel grayscale image.

[0026] Preferably, step 4: extracting the tunnel annular joint image based on the brightness value of each pixel in the tunnel grayscale image, includes:

[0027] Step 4.1: Binarize the grayscale image of the tunnel to obtain a binarized tunnel image;

[0028] Step 4.2: Extract the top image of the tunnel by cropping the binarized tunnel image according to the preset row coordinate interval;

[0029] Step 4.3: Traverse a column of the tunnel top image to find the first pixel with a brightness of 0 and the second pixel with a brightness of 0.

[0030] Step 4.4: Determine whether the difference between the row coordinates of the first pixel and the second pixel is greater than the set threshold;

[0031] Step 4.5: If the difference is greater than the set threshold, set the brightness of the corresponding column pixels in the tunnel top image to 1;

[0032] Step 4.6: If the difference is less than the set threshold, the third pixel with a brightness of 0 will be found as the first pixel, the fourth pixel with a brightness of 0 will be found as the second pixel, and the process will return to step 4.4 until every pixel on the top image of the tunnel is traversed to obtain the tunnel annular seam image.

[0033] Preferably, step 5: denoising the tunnel annular joint image to obtain a denoised tunnel annular joint image includes:

[0034] Step 5.1: Select a preset number of maximum difference points along the x-axis of the tunnel annular joint image;

[0035] Step 5.2: Use the average coordinates of all points with the largest difference as the coordinates of the center point;

[0036] Step 5.3: Calculate the distance from each point of maximum difference to the center point; the distance calculation formula is:

[0037]

[0038]

[0039]

[0040] Where D is the distance from the point of maximum difference to the center point, n is the number of points of maximum difference, and x i z is the x-coordinate of the point with the maximum difference. i The ordinate of the point with the largest difference;

[0041] Step 5.4: Remove points whose distances are not within the preset range to obtain the denoised tunnel ring seam image.

[0042] The present invention also provides a system for detecting inter-ring misalignment in shield tunnels, comprising:

[0043] The point cloud acquisition module is used to acquire the three-dimensional point cloud data of the tunnel.

[0044] The two-dimensional point cloud conversion module is used to project the three-dimensional point cloud data onto a standard cylinder and obtain two-dimensional planar point cloud data according to the unfolded shape of the standard cylinder;

[0045] A grayscale conversion module is used to convert the two-dimensional planar point cloud data into a tunnel grayscale image;

[0046] The tunnel annular joint extraction module is used to extract the tunnel annular joint image based on the brightness value of each pixel in the tunnel grayscale image.

[0047] The point cloud denoising module is used to denoise the tunnel annular joint image to obtain a denoised tunnel annular joint image.

[0048] The misalignment detection module is used to obtain the inter-ring misalignment of the tunnel based on the denoised tunnel annular joint image.

[0049] Preferably, the two-dimensional point cloud conversion module includes:

[0050] A projection unit is used to project the point cloud data onto the standard design cross-section of the tunnel, centered on the center point of the fitted ellipse, to obtain the point cloud data projection points; wherein, the cross-section of the standard cylinder is the standard design cross-section of the tunnel; the point cloud data projection points are:

[0051]

[0052]

[0053] Where P' = (x', z'), P' is the projection point of the point cloud data, R is the tunnel radius, P(x, z) is any point cloud on the tunnel, and O(x... o , z o () represents the center point of the fitted ellipse. It is the positive unit vector along the Z-axis;

[0054] A conversion unit, configured to obtain two-dimensional planar point cloud data based on the projection points of the point cloud data, includes:

[0055] Formula used:

[0056]

[0057]

[0058] Two-dimensional planar point cloud data is obtained; where ∠P'OZ is and The included angle, P”=(x”,z”) is the point after P' is expanded, x” is the x-coordinate of point P” on the two-dimensional plane point cloud, z” is the y-coordinate of point P” on the two-dimensional plane point cloud, R is the tunnel radius, x’ is the x-axis coordinate of the projection point P’, x o To fit the x-coordinate of the center point of the ellipse, z o The ordinate of the center point of the fitted ellipse is given.

[0059] Preferably, the grayscale conversion module includes:

[0060] A grid division unit is used to divide the two-dimensional planar point cloud data into grids of a preset length;

[0061] The pixel conversion unit is used to obtain a tunneled grayscale image by taking the average intensity information of all point clouds in each grid as the pixel value of the corresponding grid.

[0062] Preferably, the tunnel circumferential joint extraction module includes:

[0063] A binarization unit is used to perform binarization processing on the tunnel grayscale image to obtain a binarized tunnel image;

[0064] The cropping unit is used to crop the binarized tunnel image according to a preset row coordinate interval to obtain the top image of the tunnel;

[0065] A traversal unit is used to traverse a column of the tunnel top image to find the first pixel with a brightness of 0 and the second pixel with a brightness of 0 in sequence.

[0066] The judgment unit is used to determine whether the difference between the row coordinates of the first pixel and the second pixel is greater than a set threshold.

[0067] The first judgment result generation unit is used to set the brightness of the corresponding column pixels of the tunnel top image to 1 when the difference is greater than the set threshold.

[0068] The second judgment result generation unit is used to find the third pixel with a brightness of 0 as the first pixel and the fourth pixel with a brightness of 0 as the second pixel when the difference is less than the set threshold, and return to step 4.4 until every pixel on the top image of the tunnel is traversed to obtain the tunnel annular seam image.

[0069] Preferably, the point cloud denoising module includes:

[0070] The difference point acquisition unit is used to select a preset number of maximum difference points in the x-axis direction of the tunnel annular joint image;

[0071] The center point coordinate calculation unit is used to take the average of the coordinates of all points with the largest difference as the center point coordinates.

[0072] The distance calculation unit is used to calculate the distance from each point of maximum difference to the center point; the distance calculation formula is as follows:

[0073]

[0074]

[0075]

[0076] Where D is the distance from the point of maximum difference to the center point, n is the number of points of maximum difference, and x i z is the x-coordinate of the point with the maximum difference. i The ordinate of the point with the largest difference;

[0077] The denoising unit is used to remove points whose distances are not within a preset range to obtain a denoised tunnel circumferential seam image.

[0078] The beneficial effects of the method and system for detecting inter-ring misalignment in shield tunnels provided by this invention are as follows: Compared with the prior art, the method for detecting inter-ring misalignment in shield tunnels of this invention includes: transforming the three-dimensional point cloud data of the tunnel into two-dimensional planar point cloud data; converting the two-dimensional planar point cloud data into a tunnel grayscale image; extracting the tunnel annular joint image based on the brightness value of each pixel in the tunnel grayscale image; denoising the tunnel annular joint image to obtain a denoised tunnel annular joint image; and obtaining the amount of inter-ring misalignment in the tunnel based on the denoised tunnel annular joint image. This invention, by converting the three-dimensional point cloud data of the tunnel into a tunnel grayscale image and obtaining the amount of inter-ring misalignment based on this image, can quickly and comprehensively collect data inside the tunnel, greatly improving the detection efficiency of inter-ring misalignment in tunnels. Attached Figure Description

[0079] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art 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.

[0080] Figure 1 A flowchart of a method for detecting inter-ring misalignment in a shield tunnel provided by an embodiment of the present invention;

[0081] Figure 2 A flowchart of the three-dimensional point cloud data acquisition process for a tunnel provided in an embodiment of the present invention;

[0082] Figure 3 This is a schematic diagram of point cloud unfolding provided in an embodiment of the present invention;

[0083] Figure 4 This is a schematic diagram of point cloud rasterization provided in an embodiment of the present invention;

[0084] Figure 5 The grayscale point cloud image provided in the embodiments of the present invention;

[0085] Figure 6 The binarized tunnel image provided in this embodiment of the invention;

[0086] Figure 7 This is a schematic diagram of the tunnel top provided in an embodiment of the present invention;

[0087] Figure 8 This is a schematic diagram of a tunnel annular joint provided in an embodiment of the present invention;

[0088] Figure 9 This is a schematic diagram of the point cloud before denoising provided in an embodiment of the present invention;

[0089] Figure 10 This is a schematic diagram of the denoised point cloud provided in an embodiment of the present invention;

[0090] Figure 11 This is a schematic diagram of the analysis results of inter-ring misalignment provided in an embodiment of the present invention. Detailed Implementation

[0091] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0092] The purpose of this invention is to provide a method and system for detecting inter-ring misalignment in shield tunnels, aiming to solve the problem of low detection efficiency in existing detection methods.

[0093] Please refer to the following: Figure 1-2 To achieve the above objectives, the technical solution adopted by the present invention is: a method for detecting inter-ring misalignment in a shield tunnel, comprising the following steps:

[0094] Step 1: Obtain the 3D point cloud data of the tunnel;

[0095] In practical applications, this invention can utilize the Amberg GRP5000 mobile scanning system to acquire 3D point cloud data of shield tunnels. The Amberg GRP5000 mobile scanning system is a tunnel inspection device integrating multiple sensor technologies. This device integrates a laser scanner, odometer, sensors, a portable computer, and a power supply into a mobile trolley. During manual advancement over the tunnel, the 3D laser scanner emits a laser beam that scans the tunnel in a spiral pattern. By comparing the phase difference of the laser signals, the geometric coordinates and reflection intensity of each point on the tunnel's inner surface are obtained, thereby generating a tunnel point cloud.

[0096] During manual advancement in a tunnel, the Amberg GRP5000 mobile scanning system uses a 3D laser scanner to scan the tunnel in a spiral pattern, comparing the phase difference of the laser signals to obtain the geometric coordinates and reflection intensity of points on the tunnel's inner surface, thereby generating a tunnel point cloud. The specific steps are as follows:

[0097] (1) Instrument assembly. Assemble the various components of the Amberg GRP5000 on the subway track. After assembling the components, check and test the sensitivity and safety of the complete instrument.

[0098] (2) Parameter settings. Before scanning, a new project needs to be created in Amberg Rail. Parameter settings include the communication port of the GRP vehicle, the scanner port, configuring project attributes, customer information, data storage location, setting the boundary model and projection model, and datasets such as plan profile, longitudinal profile, design superelevation, and design rules. Then, set the basic parameters such as the working direction and odometer.

[0099] (3) Calibrate the mileage. Turn on the instrument, align the laser with the starting mileage position, and manually input the starting coordinates. Then rotate the instrument 180° and repeat the above steps to calibrate the mileage. After calibration, scanning can begin;

[0100] (4) Tunnel scanning. The surveyor controls the speed to push the trolley forward for scanning within 0.5 m / s. At the same time, the scanned tunnel lining surface, track gauge, mileage and inclination angle can be viewed on the industrial computer.

[0101] Step 2: Project the three-dimensional point cloud data onto a standard cylinder, and obtain two-dimensional planar point cloud data according to the unfolded shape of the standard cylinder;

[0102] Please see Figure 3 Furthermore, to obtain a grayscale image of the tunnel's inner surface, the 3D point cloud should first be transformed into a 2D planar point cloud. Then, the tunnel point cloud is projected onto a standard cylinder, and finally, the image is unfolded according to the standard cylinder. The specific unfolding steps are as follows:

[0103] Centered on the center point of the fitted ellipse, the point cloud data is projected onto the standard design cross section of the tunnel to obtain the point cloud data projection points; wherein, the cross section of the standard cylinder is the standard design cross section of the tunnel.

[0104] The projection points of the point cloud data are:

[0105]

[0106]

[0107] Where P' = (x', z'), P' is the projection point of the point cloud data, R is the tunnel radius, P(x, z) is any point cloud on the tunnel, and O(x... o , z o () represents the center point of the fitted ellipse. It is the positive unit vector along the Z-axis;

[0108] Two-dimensional planar point cloud data is obtained from the projection points of the point cloud data, including:

[0109] Formula used:

[0110]

[0111]

[0112] Two-dimensional planar point cloud data is obtained; where ∠P'OZ is and The included angle, P”=(x”,z”) is the point after P' is expanded, x” is the x-coordinate of point P” on the two-dimensional plane point cloud, z” is the y-coordinate of point P” on the two-dimensional plane point cloud, R is the tunnel radius, x’ is the x-axis coordinate of the projection point P’, x o To fit the x-coordinate of the center point of the ellipse, Z o The ordinate of the center point of the fitted ellipse is given.

[0113] Step 3: Convert the two-dimensional planar point cloud data into a tunnel grayscale image;

[0114] Furthermore, step 3 includes:

[0115] Step 3.1: Divide the two-dimensional planar point cloud data into grids of a preset length;

[0116] Step 3.2: Use the mean value of the intensity information of all point clouds in each grid as the pixel value of the corresponding grid to obtain the tunnel grayscale image.

[0117] Specifically, after obtaining the two-dimensional planar point cloud, it needs to be converted into image data. The image data is essentially a matrix, where each pixel has a defined row and column position and color value. Therefore, the unfolded two-dimensional point cloud is divided into grids of length l. Figure 4 This is a rasterization diagram, and the raster row and column values ​​are determined according to the following formula:

[0118] The number of vertical grid cells, m, is:

[0119]

[0120] The number of horizontal grid cells n is:

[0121]

[0122] The image data is matrix T. mn The matrix has m rows and n columns.

[0123] The mean value of the point cloud intensity information S' within a raster is used as the pixel value of that raster, that is, the point cloud depth information corresponding to the raster pixel, thus completing the conversion from point cloud to image. Figure 5 As shown:

[0124]

[0125] Where T ij S' represents the pixel value at position (i,j) in the image matrix. ij is the depth value of the point cloud located in the (i,j) grid, and n is the number of point clouds in the (i,j) grid.

[0126] Step 4: Extract the tunnel annular joint image based on the brightness value of each pixel in the tunnel grayscale image;

[0127] Furthermore, step 4 includes:

[0128] Step 4.1: Binarize the grayscale image of the tunnel to obtain a binarized tunnel image;

[0129] Step 4.2: Extract the top image of the tunnel by cropping the binarized tunnel image according to the preset row coordinate interval;

[0130] Step 4.3: Traverse a column of the tunnel top image to find the first pixel with a brightness of 0 and the second pixel with a brightness of 0.

[0131] Step 4.4: Determine whether the difference between the row coordinates of the first pixel and the second pixel is greater than the set threshold;

[0132] Step 4.5: If the difference is greater than the set threshold, set the brightness of the corresponding column pixels in the tunnel top image to 1;

[0133] Step 4.6: If the difference is less than the set threshold, the third pixel with a brightness of 0 will be found as the first pixel, the fourth pixel with a brightness of 0 will be found as the second pixel, and the process will return to step 4.4 until every pixel on the top image of the tunnel is traversed to obtain the tunnel annular seam image.

[0134] The steps for extracting tunnel circumferential joint images according to the present invention are described in detail below with reference to specific embodiments:

[0135] For the grayscale image of the tunnel's inner surface, it is necessary to extract the annular seams. Grayscale values ​​are important features for distinguishing different objects; binary segmentation can amplify the feature information of different objects within the tunnel. Figure 6 As an example of binarization, since there are no interfering objects such as pipes or supports at the top of the tunnel, and the column coordinates of a single inter-ring seam are the same in the grayscale image of the tunnel, the image of the tunnel top is cropped for inter-ring seam identification. Figure 7 This is a schematic diagram of the tunnel top. The specific steps are as follows:

[0136] The raster in the image data of the tunnel's inner surface is matrix T mn Therefore, we only need to determine the row coordinates of the top image of the tunnel that we need to extract:

[0137] a≤m≤b

[0138] Where m is the row coordinate of the image data, a is the minimum row coordinate of the image at the top of the tunnel, and b is the maximum row coordinate of the image at the top of the tunnel.

[0139] Although the captured tunnel top image is free from interference from pipes, supports, and other obstructions, the point cloud reflection intensity is affected by factors such as scanning parameters, atmospheric conditions, and scanning distance. This results in inherent interference in the captured tunnel top image due to the scanning process itself (e.g., black areas on some tunnel segments). Furthermore, bolt holes are present on the tunnel top segments. Therefore, it is necessary to extract the inter-ring seams from these interferences. The inter-ring seams are continuous vertical lines. Although the seam may be partially missing due to the scanning fieldwork, the missing portion is relatively short. Therefore, it is only necessary to determine if the difference in row coordinates of two consecutive black pixels in the same column of the image data is less than a set threshold. Figure 8 Extracting an example of a circular seam, the specific steps are as follows (this embodiment uses the first column as an example):

[0140] (1) Traverse the first column and find the pixel with a brightness of 0 (the black pixel in the figure). When the first pixel with a brightness of 0 is found, record the row coordinate of the pixel. Then find the next pixel with a brightness of 0 and record the row coordinate of the pixel.

[0141] (2) Calculate the difference between two of the recorded row coordinates and determine whether the difference l is greater than the set threshold. If it is less than the set threshold, proceed to step (3); if it is greater than the set threshold, proceed to step (4).

[0142] (3) Find the next set of pixels with a brightness of 0, record the row coordinates of the pixel, and repeat step (2).

[0143] (4) This column is not a ring seam between rings. End the search for pixels in this column and set the brightness of all pixels in this column to 1 (white image in the figure);

[0144] (5) Repeat steps (1)(2)(3)(4) for the other columns in the image.

[0145] After the above steps, the inter-ring joints of the tunnel can be extracted. Due to the influence of scanning methods and hardware, some of the identified inter-ring joints may be missing, so it is necessary to fill in the missing joints. The width of the tunnel segment in one ring is the same in the same tunnel, so the missing inter-ring joints can be filled in according to the width of the tunnel segment.

[0146] Therefore, the number of grid cells in one ring in the image is:

[0147]

[0148] In the formula, l is the grid width and L is the width of a ring segment.

[0149] The steps for repairing missing inter-ring seams in this invention are as follows:

[0150] (1) Calculate the distance k between adjacent ring joints;

[0151] (2) Calculate the number of missing rings in the inter-ring joints between adjacent rings, i.e.:

[0152]

[0153] (3) Based on the number of missing segments and the width of the segments in step (2), fill in the missing inter-ring seams.

[0154] Step 5: Denoise the tunnel annular joint image to obtain a denoised tunnel annular joint image;

[0155] Furthermore, step 5 includes:

[0156] Step 5.1: Select a preset number of maximum difference points along the x-axis of the tunnel annular joint image;

[0157] Step 5.2: Use the average coordinates of all points with the largest difference as the coordinates of the center point;

[0158] Step 5.3: Calculate the distance from each point of maximum difference to the center point; the distance calculation formula is:

[0159]

[0160] Where D is the distance from the point of maximum difference to the center point, n is the number of points of maximum difference, and x i z is the x-coordinate of the point with the maximum difference. i The ordinate of the point with the largest difference is y.

[0161] Step 5.4: Remove points whose distances are not within the preset range to obtain the denoised tunnel ring seam image.

[0162] Please see Figure 9-10 Before performing misalignment analysis, the point cloud needs to be denoised to improve the effect and accuracy of subsequent processing. Therefore, this invention uses a distance threshold to remove redundant noise points. Since the tunnel top is covered with a contact wire and the bottom is laid with a track, there are relatively many noise points at these two locations, and the laser from the 3D laser scanner is not penetrating. Therefore, n points of maximum and minimum distance in the X-axis direction are selected (i.e., the maximum distance between the maximum and minimum points in the X-axis direction, also known as the maximum difference point). These selected points are denoted as (p1, p2, ..., p...). i By calculating the average coordinates of the selected points, the coordinates of the center point of the cross section can be roughly obtained. The center of the cross section point cloud is then translated to the origin of the coordinate system, i.e., center point registration. The distance D from the point cloud to the center point is calculated. If D is between the two set thresholds, the point is considered to be a tunnel segment point cloud.

[0163]

[0164]

[0165]

[0166] Where, x c z is the x-coordinate of the center point of the cross section. c x is the ordinate of the center point of the cross section. i To extract the x-coordinate of the point cloud, Z i To extract the ordinate of the point cloud, D is the distance from the point cloud to the center point.

[0167] Step 6: Obtain the inter-ring misalignment of the tunnel based on the denoised tunnel annular joint image.

[0168] Further, step 6 includes: extracting cross-sections 8 cm to the left and right of each circumferential joint mileage as comparison cross-sections for inter-circumferential misalignment. The cross-sections are divided into 36 intervals at 10° angles. The point cloud of the cross-sections is rotated so that the track center at the bottom of the tunnel is at 0° and 360°. The average value of the point cloud at the center point of each interval is taken. Intervals at the same angle of the two cross-sections are compared to determine the misalignment amount of that interval. The results are as follows: Figure 11 As shown in the figure, the allowable misalignment value in the "Code for Construction and Acceptance of Shield Tunneling" is <15mm. Obviously, the 150°-160° in the figure exceeds 15mm, so it needs to be inspected and repaired.

[0169] This invention acquires point cloud data through a mobile scanning system and obtains tunnel circumferential joint images based on this data. It can quickly extract the location of the circumferential joint. Compared with manual caliper measurement, this invention is more efficient and has a more comprehensive detection range.

[0170] The present invention also provides a system for detecting inter-ring misalignment in shield tunnels, comprising:

[0171] The point cloud acquisition module is used to acquire the three-dimensional point cloud data of the tunnel.

[0172] The two-dimensional point cloud conversion module is used to project the three-dimensional point cloud data onto a standard cylinder and obtain two-dimensional planar point cloud data according to the unfolded shape of the standard cylinder;

[0173] A grayscale conversion module is used to convert the two-dimensional planar point cloud data into a tunnel grayscale image;

[0174] The tunnel annular joint extraction module is used to extract the tunnel annular joint image based on the brightness value of each pixel in the tunnel grayscale image.

[0175] The point cloud denoising module is used to denoise the tunnel annular joint image to obtain a denoised tunnel annular joint image.

[0176] The misalignment detection module is used to obtain the inter-ring misalignment of the tunnel based on the denoised tunnel annular joint image.

[0177] Preferably, the two-dimensional point cloud conversion module includes:

[0178] A projection unit is used to project the point cloud data onto the standard design cross-section of the tunnel, centered on the center point of the fitted ellipse, to obtain the point cloud data projection points; wherein, the cross-section of the standard cylinder is the standard design cross-section of the tunnel; the point cloud data projection points are:

[0179]

[0180]

[0181] Where P' = (x′', z′), P' is the projection point of the point cloud data, R is the tunnel radius, P(x, z) is any point cloud on the tunnel, and O(x′', z′') is the distance from the tunnel to the tunnel. o , z o () represents the center point of the fitted ellipse. It is the positive unit vector along the Z-axis;

[0182] A conversion unit, configured to obtain two-dimensional planar point cloud data based on the projection points of the point cloud data, includes:

[0183] Formula used:

[0184]

[0185]

[0186] Two-dimensional planar point cloud data is obtained; where ∠P'OZ is and The included angle, P”=(x”,z”) is the point after P' is expanded, x” is the x-coordinate of point P” on the two-dimensional plane point cloud, z” is the y-coordinate of point P” on the two-dimensional plane point cloud, R is the tunnel radius, x’ is the x-axis coordinate of the projection point P’, x o To fit the x-coordinate of the center point of the ellipse, z o The ordinate of the center point of the fitted ellipse is given.

[0187] Preferably, the grayscale conversion module includes:

[0188] A grid division unit is used to divide the two-dimensional planar point cloud data into grids of a preset length;

[0189] The pixel conversion unit is used to obtain a tunneled grayscale image by taking the average intensity information of all point clouds in each grid as the pixel value of the corresponding grid.

[0190] Preferably, the tunnel circumferential joint extraction module includes:

[0191] A binarization unit is used to perform binarization processing on the tunnel grayscale image to obtain a binarized tunnel image;

[0192] The cropping unit is used to crop the binarized tunnel image according to a preset row coordinate interval to obtain the top image of the tunnel;

[0193] A traversal unit is used to traverse a column of the tunnel top image to find the first pixel with a brightness of 0 and the second pixel with a brightness of 0 in sequence.

[0194] The judgment unit is used to determine whether the difference between the row coordinates of the first pixel and the second pixel is greater than a set threshold.

[0195] The first judgment result generation unit is used to set the brightness of the corresponding column pixels of the tunnel top image to 1 when the difference is greater than the set threshold.

[0196] The second judgment result generation unit is used to find the third pixel with a brightness of 0 as the first pixel and the fourth pixel with a brightness of 0 as the second pixel when the difference is less than the set threshold, and return to step 4.4 until every pixel on the top image of the tunnel is traversed to obtain the tunnel annular seam image.

[0197] Preferably, the point cloud denoising module includes:

[0198] The difference point acquisition unit is used to select a preset number of maximum difference points in the x-axis direction of the tunnel annular joint image;

[0199] The center point coordinate calculation unit is used to take the average of the coordinates of all points with the largest difference as the center point coordinates.

[0200] The distance calculation unit is used to calculate the distance from each point of maximum difference to the center point; the distance calculation formula is as follows:

[0201]

[0202]

[0203]

[0204] Where D is the distance from the point of maximum difference to the center point, n is the number of points of maximum difference, and x i z is the x-coordinate of the point with the maximum difference. i The ordinate of the point with the largest difference;

[0205] The denoising unit is used to remove points whose distances are not within a preset range to obtain a denoised tunnel circumferential seam image.

[0206] The beneficial effects of the method and system for detecting inter-ring misalignment in shield tunnels provided by this invention are as follows: Compared with the prior art, this invention converts the three-dimensional point cloud data of the tunnel into a grayscale image of the tunnel, and obtains the amount of inter-ring misalignment based on this image. This allows for rapid and comprehensive acquisition of data inside the tunnel, greatly improving the detection efficiency of inter-ring misalignment in tunnels.

[0207] 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, and improvements 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 inter-ring misalignment in a shield tunnel, characterized in that, Includes the following steps: Step 1: Obtain the 3D point cloud data of the tunnel; Step 2: Project the three-dimensional point cloud data onto a standard cylinder, and obtain two-dimensional planar point cloud data according to the unfolded shape of the standard cylinder; Step 3: Convert the two-dimensional planar point cloud data into a tunnel grayscale image; Step 4: Extract the tunnel annular joint image based on the brightness value of each pixel in the tunnel grayscale image; Step 5: Denoise the tunnel annular joint image to obtain a denoised tunnel annular joint image; Step 6: Obtain the inter-ring misalignment of the tunnel based on the denoised tunnel annular joint image; Step 4: Extracting the tunnel annular joint image based on the brightness value of each pixel in the tunnel grayscale image, including: Step 4.1: Binarize the grayscale image of the tunnel to obtain a binarized tunnel image; Step 4.2: Extract the top image of the tunnel by cropping the binarized tunnel image according to the preset row coordinate interval; Step 4.3: Traverse a column of the tunnel top image to find the first pixel with a brightness of 0 and the second pixel with a brightness of 0. Step 4.4: Determine whether the difference between the row coordinates of the first pixel and the second pixel is greater than the set threshold; Step 4.5: If the difference is greater than the set threshold, set the brightness of the corresponding column pixels in the tunnel top image to 1; Step 4.6: If the difference is less than the set threshold, the third pixel with a brightness of 0 will be found as the first pixel, the fourth pixel with a brightness of 0 will be found as the second pixel, and the process will return to step 4.4 until every pixel on the top image of the tunnel is traversed to obtain the tunnel annular seam image. Step 5: Denoising the tunnel annular joint image to obtain a denoised tunnel annular joint image, including: Step 5.1: Select a preset number of maximum difference points along the x-axis of the tunnel annular joint image; Step 5.2: Use the average coordinates of all points with the largest difference as the coordinates of the center point; Step 5.3: Calculate the distance from each point of maximum difference to the center point; the distance calculation formula is: Where D is the distance from the point of maximum difference to the center point, n is the number of points of maximum difference, and x i z is the x-coordinate of the point with the maximum difference. i The ordinate of the point with the largest difference; Step 5.4: Remove points whose distances are not within the preset range to obtain the denoised tunnel annular joint image; Step 6: Obtaining the inter-ring misalignment of the tunnel based on the denoised tunnel annular joint image includes: extracting cross-sections 8cm to the left and right of each annular joint mileage as comparison cross-sections for inter-ring misalignment; dividing the cross-sections into 36 intervals every 10° angle; rotating the cross-section point cloud so that the track center at the bottom of the tunnel is 0° and 360°; taking the average value of the center point of the point cloud of each interval; comparing the intervals of the same angle of the two cross-sections to obtain the misalignment of that interval.

2. The method for detecting inter-ring misalignment in a shield tunnel as described in claim 1, characterized in that, Step 2: Projecting the three-dimensional point cloud data onto a standard cylinder, and obtaining two-dimensional planar point cloud data according to the unfolded shape of the standard cylinder, including: Centered on the center point of the fitted ellipse, the point cloud data is projected onto the standard design cross-section of the tunnel to obtain the point cloud data projection points; wherein, the cross-section of the standard cylinder is the standard design cross-section of the tunnel; the point cloud data projection points are: Where P′=(x′,z′), P′ is the projection point of the point cloud data, R is the tunnel radius, P(x,z) is any point cloud on the tunnel, and O(x o , z o () represents the center point of the fitted ellipse. It is the positive unit vector along the Z-axis; Two-dimensional planar point cloud data is obtained from the projection points of the point cloud data, including: Formula used: Two-dimensional planar point cloud data is obtained; where ∠P′OZ is and The included angle, P”=(x”,z”) is the point after P′ is expanded, x” is the x-coordinate of point P” on the two-dimensional plane point cloud, z” is the y-coordinate of point P” on the two-dimensional plane point cloud, R is the tunnel radius, x′ is the x-axis coordinate of the projection point P′, x o To fit the x-coordinate of the center point of the ellipse, z o The ordinate of the center point of the fitted ellipse is given.

3. The method for detecting inter-ring misalignment in a shield tunnel as described in claim 1, characterized in that, Step 3: Converting the two-dimensional planar point cloud data into a tunnel grayscale image, including: Step 3.1: Divide the two-dimensional planar point cloud data into grids of a preset length; Step 3.2: Use the mean value of the intensity information of all point clouds in each grid as the pixel value of the corresponding grid to obtain the tunnel grayscale image.

4. A system for detecting inter-ring misalignment in a shield tunnel, used to implement the steps included in the method for detecting inter-ring misalignment in a shield tunnel according to any one of claims 1-3, characterized in that, include: The point cloud acquisition module is used to acquire the three-dimensional point cloud data of the tunnel. The two-dimensional point cloud conversion module is used to project the three-dimensional point cloud data onto a standard cylinder and obtain two-dimensional planar point cloud data according to the unfolded shape of the standard cylinder; A grayscale conversion module is used to convert the two-dimensional planar point cloud data into a tunnel grayscale image; The tunnel annular joint extraction module is used to extract the tunnel annular joint image based on the brightness value of each pixel in the tunnel grayscale image. The point cloud denoising module is used to denoise the tunnel annular joint image to obtain a denoised tunnel annular joint image. The misalignment detection module is used to obtain the inter-ring misalignment of the tunnel based on the denoised tunnel annular joint image.

5. The inter-ring misalignment detection system for shield tunnels as described in claim 4, characterized in that, The two-dimensional point cloud conversion module includes: A projection unit is used to project the point cloud data onto the standard design cross-section of the tunnel, centered on the center point of the fitted ellipse, to obtain the point cloud data projection points; wherein, the cross-section of the standard cylinder is the standard design cross-section of the tunnel; the point cloud data projection points are: Where P′=(x′,z′), P′ is the projection point of the point cloud data, R is the tunnel radius, P(x,z) is any point cloud on the tunnel, and O(x o , z o () represents the center point of the fitted ellipse. It is the positive unit vector along the Z-axis; A conversion unit, configured to obtain two-dimensional planar point cloud data based on the projection points of the point cloud data, includes: Formula used: Two-dimensional planar point cloud data is obtained; where ∠P′OZ is and The included angle, P”=(x”,z”) is the point after P′ is expanded, x” is the x-coordinate of point P” on the two-dimensional plane point cloud, z” is the y-coordinate of point P” on the two-dimensional plane point cloud, R is the tunnel radius, x′ is the x-axis coordinate of the projection point P′, x o To fit the x-coordinate of the center point of the ellipse, z o The ordinate of the center point of the fitted ellipse is given.

6. The inter-ring misalignment detection system for shield tunnels as described in claim 4, characterized in that, The grayscale conversion module includes: A grid division unit is used to divide the two-dimensional planar point cloud data into grids of a preset length; The pixel conversion unit is used to obtain a tunneled grayscale image by taking the average intensity information of all point clouds in each grid as the pixel value of the corresponding grid.

7. The inter-ring misalignment detection system for shield tunnels as described in claim 4, characterized in that, The tunnel circumferential joint extraction module includes: A binarization unit is used to perform binarization processing on the tunnel grayscale image to obtain a binarized tunnel image; The cropping unit is used to crop the binarized tunnel image according to a preset row coordinate interval to obtain the top image of the tunnel; A traversal unit is used to traverse a column of the tunnel top image to find the first pixel with a brightness of 0 and the second pixel with a brightness of 0 in sequence. The judgment unit is used to determine whether the difference between the row coordinates of the first pixel and the second pixel is greater than a set threshold. The first judgment result generation unit is used to set the brightness of the corresponding column pixels of the tunnel top image to 1 when the difference is greater than the set threshold. The second judgment result generation unit is used to find the third pixel with a brightness of 0 as the first pixel and the fourth pixel with a brightness of 0 as the second pixel when the difference is less than the set threshold, and return to step 4.4 until every pixel on the top image of the tunnel is traversed to obtain the tunnel annular seam image.

8. The inter-ring misalignment detection system for shield tunnels as described in claim 4, characterized in that, The point cloud denoising module includes: The difference point acquisition unit is used to select a preset number of maximum difference points in the x-axis direction of the tunnel annular joint image; The center point coordinate calculation unit is used to take the average of the coordinates of all points with the largest difference as the center point coordinates. The distance calculation unit is used to calculate the distance from each point of maximum difference to the center point; the distance calculation formula is as follows: Where D is the distance from the point of maximum difference to the center point, n is the number of points of maximum difference, and x i z is the x-coordinate of the point with the maximum difference. i The ordinate of the point with the largest difference; The denoising unit is used to remove points whose distances are not within a preset range to obtain a denoised tunnel circumferential seam image.

Images