Deformation monitoring method, system, equipment and medium based on shield tunnel point cloud

Abstract

The present invention discloses a deformation monitoring method, system, equipment and medium based on shield tunnel point cloud, which significantly improves the accuracy of tunnel deformation monitoring through denoising and expansion operations, and effectively reduces external interference. At the same time, it has the ability to quickly process a large amount of point cloud data for real-time monitoring and early warning. Through the set deformation threshold, the method can automatically identify the deformation area, reduce manual intervention, and improve monitoring efficiency. In addition, the method is not only suitable for deformation monitoring of shield tunnels, but can also be extended to the monitoring of other types of tunnels or underground structures.

Description

Technical Field

[0001] The present invention relates to the field of mobile sensing technology, and more specifically, to a deformation monitoring method, system, equipment and medium based on shield tunnel point cloud. Background Art

[0002] With the continuous development of tunnel engineering, especially the widespread application of shield tunnels, how to accurately monitor the deformation of tunnels and ensure the safety and stability of tunnel structures has become an important task in engineering construction. Traditional tunnel deformation monitoring methods, such as total stations and laser rangefinders, are limited by monitoring accuracy and real-time performance, making it difficult to efficiently and comprehensively monitor the deformation of the entire tunnel.

[0003] In recent years, with the rapid development of laser scanning technology and point cloud data processing technology, the use of point cloud data for tunnel deformation monitoring has gradually become an effective means. Point cloud data often contains noise, which affects the accuracy of the data and thus affects the accuracy of the deformation monitoring results. Low efficiency of data processing and analysis: Existing methods usually rely on a large amount of manual operations when processing tunnel point cloud data, and the processing efficiency is low, and deformation monitoring results cannot be given in real time. At present, many methods still have not established an automatic recognition mechanism for deformation thresholds based on point clouds, resulting in inaccurate determination of deformation areas and easy interference from external environmental factors; through point cloud scanning, three-dimensional spatial data inside and outside the tunnel can be obtained, and then accurate monitoring of tunnel deformation can be performed by analyzing the changes in point clouds. Summary of the invention

[0004] In order to overcome the defects of low accuracy of deformation monitoring results and low efficiency of data processing and deformation analysis in the prior art, the present invention provides a deformation monitoring method, system, equipment and medium based on shield tunnel point cloud, which can quickly and accurately detect the deformation area of ​​the tunnel through denoising, expansion, deformation comparison and area identification of point cloud data, thereby improving the efficiency and accuracy of monitoring.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] The present invention provides a deformation monitoring method based on shield tunnel point cloud, comprising:

[0007] Scan the tunnel multiple times to obtain reference tunnel point cloud data and test tunnel point cloud data at different times;

[0008] De-noising the reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods to obtain the de-noised reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods;

[0009] Based on the denoised reference tunnel point cloud data and the measured tunnel point cloud data at different periods, the data are expanded according to the designed center line of the tunnel to obtain the corresponding reference two-dimensional grid depth array and the measured two-dimensional grid depth array at different periods;

[0010] Compare the depth values ​​in the corresponding reference two-dimensional grid depth array with the depth values ​​in the two-dimensional grid depth array to be tested at different periods to obtain a corresponding depth difference two-dimensional array;

[0011] A depth deformation threshold and a deformation area threshold are set, and the depth value of the corresponding depth difference two-dimensional array is compared with the depth deformation threshold and the deformation area threshold, and the deformation of the tunnel is determined based on the comparison result.

[0012] Preferably, the tunnel is scanned multiple times to obtain reference tunnel point cloud data and test tunnel point cloud data at different times, including:

[0013] Use a total station to establish control points on the tunnel to be measured, and establish a coordinate system consistent with the target coordinate system;

[0014] A three-dimensional laser instrument is set up, and three prism target balls are arranged at predetermined positions from the laser instrument, wherein the first target ball is located in the direction of tunnel advancement, the second target ball is located in the opposite direction of tunnel advancement, and the third target ball is located between the two and close to the three-dimensional laser instrument;

[0015] Start the 3D laser instrument to collect raw laser data, and use the total station to collect the coordinates of the target sphere prism until the data of the entire tunnel to be measured is collected;

[0016] Perform data processing, convert the original laser data into point cloud data, find the position of the target ball, intercept the point cloud of the target ball, identify and fit the point cloud coordinates of the center of all target balls;

[0017] Record the point cloud coordinates of the center of the target sphere in each point cloud and the corresponding coordinates of the center of the target sphere prism collected by the total station as the same-name point pairs in different coordinate systems;

[0018] Using the three pairs of points with the same name in each point cloud data, the transformation matrix from each point cloud to the target coordinate system is calculated through singular value decomposition, and all point clouds are converted to the target coordinate system to obtain the reference tunnel point cloud data and the point cloud data of the tunnel to be tested at different periods.

[0019] Preferably, the reference tunnel point cloud data and the measured tunnel point cloud data at different periods are subjected to denoising processing to obtain the denoised reference tunnel point cloud data and the measured tunnel point cloud data at different periods, including:

[0020] A nearest neighbor search tree is established to process the reference tunnel point cloud data to be filtered and the tunnel point cloud data to be tested at different periods;

[0021] Traversing the point cloud data, searching for all points within a preset radius through the nearest neighbor search tree, and being able to adjust the filter sensitivity through an external interface setting;

[0022] For each point set within a preset radius, calculate the mean of all points;

[0023] Calculate the standard deviation of the distance from each point in the point set to the mean, which is used to indicate the degree of dispersion of the point distribution;

[0024] If the distance from the point to the mean is greater than the preset multiple of the standard deviation, the point is marked as a deleted point;

[0025] After the traversal is completed, all marked deleted points are removed to obtain the denoised reference tunnel point cloud data and the measured tunnel point cloud data at different periods.

[0026] Preferably, based on the denoised reference tunnel point cloud data and the measured tunnel point cloud data at different periods, the corresponding reference two-dimensional grid depth array and the measured two-dimensional grid depth array at different periods are expanded according to the design center line of the tunnel, including:

[0027] Input the corresponding design parameters according to the design drawings to obtain the design center line;

[0028] Discretize the designed centerline at specified intervals to obtain a discrete centerline point set and a corresponding mileage set;

[0029] Traverse the denoised reference tunnel point cloud data and calculate the projection points of the two points closest to the midline point set in the point cloud data on the designed midline;

[0030] Determine the first plane coordinates of the points of the corresponding point cloud data after expansion according to the indexes of the two points closest to the midline point set in the point cloud data and the mileage data in the corresponding mileage set,

[0031] Calculate the vector from the projection point of a point in the reference tunnel point cloud data on the design centerline to the point, where the modulus of the vector is the depth value of the point after expansion;

[0032] According to the vector from the projection point on the design centerline to the point in the reference tunnel point cloud data, the second plane coordinates of the point corresponding to the point cloud data after expansion are calculated by the arc cosine function;

[0033] A corresponding reference two-dimensional grid depth array is constructed according to the first plane coordinates and the second plane coordinates, and depth values ​​of points in the reference tunnel point cloud data after expansion are stored in the reference two-dimensional grid depth array.

[0034] According to the tunnel point cloud data to be tested at different periods, the two-dimensional grid depth arrays to be tested at different periods are constructed in accordance with the above steps, and the depth values ​​of the points in the tunnel point cloud data to be tested at different periods after expansion are stored in the two-dimensional grid depth arrays to be tested at different periods.

[0035] Preferably, a depth deformation threshold and a deformation area threshold are set, and the depth value of the corresponding depth difference two-dimensional array is compared with the depth deformation threshold and the deformation area threshold, and the deformation of the tunnel is determined based on the comparison result, including:

[0036] Traverse each depth value in the depth difference two-dimensional array and compare it with the set depth deformation threshold. If the depth value exceeds the depth deformation threshold, mark the area where the grid is located as a deformation area;

[0037] Expand the grid marked as deformed outwards. If the adjacent grid is also a deformed grid, connect the current deformed grid.

[0038] Count the number of connected areas of the deformed grid. If the number of connected areas is greater than the preset deformation area threshold, mark the area as a dangerous area.

[0039] Process the depth difference two-dimensional arrays corresponding to all periods, and mark the dangerous areas corresponding to all depth difference two-dimensional arrays.

[0040] Preferably, it also includes:

[0041] Comparing the depth values ​​in the corresponding reference two-dimensional grid depth array with the depth values ​​in the two-dimensional grid depth array to be tested at different periods, to obtain a number of corresponding depth difference two-dimensional arrays;

[0042] Create a color band, traverse all the depth difference two-dimensional arrays, match the corresponding color of the color band according to the depth value in each depth difference two-dimensional array, and obtain the corresponding thermal change map.

[0043] Preferably, it also includes:

[0044] Generate thermal change videos from thermal change diagrams in sequence and store them;

[0045] The deformation trend is analyzed based on the thermal change video. If there is a deformation area and the deformation area has not reached the deformation area threshold, it is judged based on historical data whether the deformation area has a trend of increasing. When the deformation area has a trend of increasing, it is marked as a potential deformation area, and the staff is notified through smart devices to conduct on-site inspections.

[0046] The present invention also provides a deformation monitoring system based on shield tunnel point cloud, which is used to implement the above method, including:

[0047] The data acquisition module scans the tunnel multiple times to obtain the reference tunnel point cloud data and the measured tunnel point cloud data at different periods;

[0048] A denoising module performs denoising on the reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods, thereby obtaining the denoised reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods;

[0049] The centerline expansion module expands the de-noised reference tunnel point cloud data and the measured tunnel point cloud data at different periods according to the designed centerline of the tunnel to obtain the corresponding reference two-dimensional grid depth array and the measured two-dimensional grid depth array at different periods;

[0050] The difference comparison module compares the depth values ​​in the corresponding reference two-dimensional grid depth array with the depth values ​​in the two-dimensional grid depth array to be tested at different periods to obtain a corresponding depth difference two-dimensional array;

[0051] The deformation determination module sets a depth deformation threshold and a deformation area threshold, compares the depth value of the corresponding depth difference two-dimensional array with the depth deformation threshold and the deformation area threshold, and determines the deformation of the tunnel based on the comparison result.

[0052] The present invention also provides an electronic device, comprising:

[0053] one or more processors;

[0054] A memory for storing one or more programs;

[0055] When the one or more programs are executed by the one or more processors, the processors are enabled to implement the above method.

[0056] The present invention also provides a computer-readable storage medium containing a computer program, on which a computer program is stored, and the program implements the above method when executed by one or more processors.

[0057] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0058] The present invention proposes a deformation monitoring method, system, equipment and medium based on shield tunnel point cloud, which significantly improves the accuracy of tunnel deformation monitoring through denoising and expansion operations, and effectively reduces external interference. At the same time, it has the ability to quickly process a large amount of point cloud data for real-time monitoring and early warning. Through the set deformation threshold, the method can automatically identify the deformation area, reduce manual intervention, and improve monitoring efficiency. In addition, the method is not only suitable for deformation monitoring of shield tunnels, but can also be extended to the monitoring of other types of tunnels or underground structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a flow chart of a deformation monitoring method based on shield tunnel point cloud described in Example 1;

[0060] Figure 2 It is a flow chart of a deformation monitoring method based on shield tunnel point cloud described in Example 2;

[0061] Figure 3 This is a structural schematic diagram of a deformation monitoring system based on shield tunnel point cloud described in Example 3. DETAILED DESCRIPTION

[0062] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0063] In order to better illustrate the present embodiment, some parts in the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product;

[0064] It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0065] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0066] Example 1

[0067] This embodiment provides a deformation monitoring method based on shield tunnel point cloud, such as Figure 1 As shown, including:

[0068] Scan the tunnel multiple times to obtain reference tunnel point cloud data and test tunnel point cloud data at different times;

[0069] De-noising the reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods to obtain the de-noised reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods;

[0070] Based on the denoised reference tunnel point cloud data and the measured tunnel point cloud data at different periods, the data are expanded according to the designed center line of the tunnel to obtain the corresponding reference two-dimensional grid depth array and the measured two-dimensional grid depth array at different periods;

[0071] Compare the depth values ​​in the corresponding reference two-dimensional grid depth array with the depth values ​​in the two-dimensional grid depth array to be tested at different periods to obtain a corresponding depth difference two-dimensional array;

[0072] A depth deformation threshold and a deformation area threshold are set, and the depth value of the corresponding depth difference two-dimensional array is compared with the depth deformation threshold and the deformation area threshold, and the deformation of the tunnel is determined based on the comparison result.

[0073] This embodiment significantly improves the accuracy of tunnel deformation monitoring through denoising and expansion operations, and effectively reduces external interference. At the same time, it has the ability to quickly process a large amount of point cloud data for real-time monitoring and early warning. Through the set deformation threshold, the method can automatically identify the deformation area, reduce manual intervention, and improve monitoring efficiency. In addition, this method is not only suitable for deformation monitoring of shield tunnels, but can also be extended to monitoring of other types of tunnels or underground structures.

[0074] Example 2

[0075] This embodiment provides a deformation monitoring method based on shield tunnel point cloud, including:

[0076] Scan the tunnel multiple times to obtain reference tunnel point cloud data and test tunnel point cloud data at different times;

[0077] De-noising the reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods to obtain the de-noised reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods;

[0078] Based on the denoised reference tunnel point cloud data and the measured tunnel point cloud data at different periods, the data are expanded according to the designed center line of the tunnel to obtain the corresponding reference two-dimensional grid depth array and the measured two-dimensional grid depth array at different periods;

[0079] Compare the depth values ​​in the corresponding reference two-dimensional grid depth array with the depth values ​​in the two-dimensional grid depth array to be tested at different periods to obtain a corresponding depth difference two-dimensional array;

[0080] A depth deformation threshold and a deformation area threshold are set, and the depth value of the corresponding depth difference two-dimensional array is compared with the depth deformation threshold and the deformation area threshold, and the deformation of the tunnel is determined based on the comparison result.

[0081] Scan the tunnel multiple times to obtain reference tunnel point cloud data and test tunnel point cloud data at different times, including:

[0082] Use a total station to establish control points on the tunnel to be measured, and establish a coordinate system consistent with the target coordinate system;

[0083] A three-dimensional laser instrument is set up, and three prism target balls are arranged at predetermined positions from the laser instrument, wherein the first target ball is located in the direction of tunnel advancement, the second target ball is located in the opposite direction of tunnel advancement, and the third target ball is located between the two and close to the three-dimensional laser instrument;

[0084] Start the 3D laser instrument to collect raw laser data, and use the total station to collect the coordinates of the target sphere prism until the data of the entire tunnel to be measured is collected;

[0085] Perform data processing, convert the original laser data into point cloud data, find the position of the target ball, intercept the point cloud of the target ball, identify and fit the point cloud coordinates of the center of all target balls;

[0086] Record the point cloud coordinates of the center of the target sphere in each point cloud and the corresponding coordinates of the center of the target sphere prism collected by the total station as the same-name point pairs in different coordinate systems;

[0087] Using the three pairs of points with the same name in each point cloud data, the transformation matrix from each point cloud to the target coordinate system is calculated through singular value decomposition, and all point clouds are converted to the target coordinate system to obtain the reference tunnel point cloud data and the point cloud data of the tunnel to be tested at different periods.

[0088] The reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods are denoised to obtain the denoised reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods, including:

[0089] A nearest neighbor search tree is established to process the reference tunnel point cloud data to be filtered and the tunnel point cloud data to be tested at different periods;

[0090] Traversing the point cloud data, searching for all points within a preset radius through the nearest neighbor search tree, and being able to adjust the filter sensitivity through an external interface setting;

[0091] For each point set within a preset radius, calculate the mean of all points;

[0092] Calculate the standard deviation of the distance from each point in the point set to the mean, which is used to indicate the degree of dispersion of the point distribution;

[0093] If the distance from the point to the mean is greater than the preset multiple of the standard deviation, the point is marked as a deleted point;

[0094] After the traversal is completed, all marked deleted points are removed to obtain the denoised reference tunnel point cloud data and the measured tunnel point cloud data at different periods.

[0095] Based on the denoised reference tunnel point cloud data and the measured tunnel point cloud data at different periods, the corresponding reference two-dimensional grid depth array and the measured two-dimensional grid depth array at different periods are obtained according to the designed center line of the tunnel, including:

[0096] Input the corresponding design parameters according to the design drawings to obtain the design center line;

[0097] Discretize the designed centerline at specified intervals to obtain a discrete centerline point set and a corresponding mileage set;

[0098] Traverse the denoised reference tunnel point cloud data and calculate the projection points of the two points closest to the midline point set in the point cloud data on the designed midline;

[0099] Determine the first plane coordinates of the points of the corresponding point cloud data after expansion according to the indexes of the two points closest to the midline point set in the point cloud data and the mileage data in the corresponding mileage set,

[0100] Calculate the vector from the projection point of a point in the reference tunnel point cloud data on the design centerline to the point, where the modulus of the vector is the depth value of the point after expansion;

[0101] According to the vector from the projection point on the design centerline to the point in the reference tunnel point cloud data, the second plane coordinates of the point corresponding to the point cloud data after expansion are calculated by the arc cosine function;

[0102] A corresponding reference two-dimensional grid depth array is constructed according to the first plane coordinates and the second plane coordinates, and depth values ​​of points in the reference tunnel point cloud data after expansion are stored in the reference two-dimensional grid depth array.

[0103] According to the tunnel point cloud data to be tested at different periods, the two-dimensional grid depth arrays to be tested at different periods are constructed in accordance with the above steps, and the depth values ​​of the points in the tunnel point cloud data to be tested at different periods after expansion are stored in the two-dimensional grid depth arrays to be tested at different periods.

[0104] A depth deformation threshold and a deformation area threshold are set, and the depth value of the corresponding depth difference two-dimensional array is compared with the depth deformation threshold and the deformation area threshold, and the deformation of the tunnel is determined based on the comparison result, including:

[0105] Traverse each depth value in the depth difference two-dimensional array and compare it with the set depth deformation threshold. If the depth value exceeds the depth deformation threshold, mark the area where the grid is located as a deformation area;

[0106] Expand the grid marked as deformed outwards. If the adjacent grid is also a deformed grid, connect the current deformed grid.

[0107] Count the number of connected areas of the deformed grid. If the number of connected areas is greater than the preset deformation area threshold, mark the area as a dangerous area.

[0108] Process the depth difference two-dimensional arrays corresponding to all periods, and mark the dangerous areas corresponding to all depth difference two-dimensional arrays.

[0109] Also includes:

[0110] Comparing the depth values ​​in the corresponding reference two-dimensional grid depth array with the depth values ​​in the two-dimensional grid depth array to be tested at different periods, to obtain a number of corresponding depth difference two-dimensional arrays;

[0111] Create a color band, traverse all the depth difference two-dimensional arrays, match the corresponding color of the color band according to the depth value in each depth difference two-dimensional array, and obtain the corresponding thermal change map.

[0112] Also includes:

[0113] Generate thermal change videos from thermal change diagrams in sequence and store them;

[0114] The deformation trend is analyzed based on the thermal change video. If there is a deformation area and the deformation area has not reached the deformation area threshold, it is judged based on historical data whether the deformation area has a trend of increasing. When the deformation area has a trend of increasing, it is marked as a potential deformation area, and the staff is notified through smart devices to conduct on-site inspections.

[0115] In a specific embodiment, a deformation monitoring method based on shield tunnel point cloud is provided. Figure 2 As shown, including:

[0116] 1. Point cloud data acquisition

[0117] Use 3D laser or other 3D scanning equipment to obtain point cloud data in the tunnel. When collecting data, set appropriate scanning intervals and accuracy to ensure the integrity and accuracy of the data.

[0118] 2. Data collection and coordinate conversion

[0119] The point cloud collected by the three-dimensional laser instrument is the current local coordinate, which needs to be converted to the target coordinate system of the tunnel design through the arranged target ball or other target objects. The present invention adopts a prism target ball (the target ball is half a smooth sphere and the other half is a prism. The center of the prism is the center of the target ball. The coordinates of the center of the prism collected by the total station and the coordinates of the center of the point cloud collected by the three-dimensional laser instrument are the same points in different coordinate systems. The two are combined into a target point pair).

[0120] 2.1 Total station establishment, through the control points on the tunnel, the coordinate system of the total station after establishment is consistent with the coordinate system on the road design. 3D laser instrument station, three prism target balls, respectively placed 3-8 meters away from the 3D laser instrument, target ball 1 is placed in the direction of the tunnel, target ball 2 is placed in the opposite direction of the tunnel, and target ball 3 is placed between the two, close to the 3D laser instrument.

[0121] 2.2 Start the 3D laser instrument to collect point cloud data, and the total station to collect the target sphere prism coordinates;

[0122] 2.3 The 3D laser instrument is spaced about 20 meters apart, and steps 1 and 2 are repeated until the complete data of the tunnel to be measured is collected;

[0123] 2.4 Data internal page processing, converting the original laser data into point cloud data, browsing through the interface, finding the location of the target ball, using the mouse to interactively cut the point cloud of the target ball, fitting the point cloud coordinates of the ball center, identifying and fitting the point cloud coordinates of all target ball centers in the point cloud;

[0124] 2.5 Follow up the corresponding relationship during collection, record the point cloud coordinates of the target sphere center in each point cloud and the corresponding prism sphere center coordinates collected by the total station, the two are the same-name point pairs in different coordinate systems;

[0125] 2.6 Using the three pairs of points of the same name in each point cloud, calculate the transformation matrix of each point cloud to the road design coordinate system (total station coordinate system, the two are consistent) through singular value decomposition (SVD), and apply it to transfer all point clouds to the road design coordinate system.

[0126] 3. Point cloud denoising

[0127] The acquired point cloud data usually contains a lot of noise and needs to be denoised. The acquired raw point cloud data is filtered and the noise points are smoothed using common denoising algorithms such as Gaussian filtering, statistical filtering, radius filtering or median filtering. The denoised point cloud data can effectively reduce the impact of external factors on data analysis and improve the accuracy of subsequent processing results. The present invention adopts statistical filtering.

[0128] 3.1 Establish a nearest neighbor search tree (KD-Tree) for the point cloud to be filtered;

[0129] 3.2 Traverse the point cloud and search all points within a preset radius of 0.2m (this radius is set through the external interface and is a filtering parameter that can adjust the sensitivity of the filter) through the nearest neighbor search tree;

[0130] 3.3 Calculate the mean of all points within the radius;

[0131] 3.4 Calculate the standard deviation: According to the distance from all points within the radius to the mean, calculate the standard deviation to indicate the degree of dispersion of the distribution of these points:

[0132] 3.5 If the passing distance is greater than 5 times the standard deviation, the point is considered to be deleted;

[0133] 3.6 After all points are traversed, the marked deletion points are uniformly removed, and the point cloud denoising is completed.

[0134] 4. Point cloud expansion and grid storage

[0135] The denoised point cloud data is expanded according to the centerline of the tunnel, and the 3D point cloud is projected onto a 2D plane using an appropriate mathematical model, retaining the depth information of each point and storing the data in a grid format. Each grid cell corresponds to a specific depth value, which is convenient for subsequent comparison and analysis.

[0136] 4.1 According to the design drawings, input the design parameters through the interface to obtain the design centerline.

[0137] 4.2 Discretize the centerline at a specified interval to obtain a discrete centerline point set DC and a mileage set DCMileage of corresponding points;

[0138] 4.3 Traverse the point cloud and find the projection point Pci of point Pi on the point cloud on the midline;

[0139] 4.3.1 Find the two points Pci0 and Pci1 that are closest to Pi in the point set DC;

[0140] 4.3.2 Find the projection point of Pi on the straight line where Pci0 and Pci1 are located, which is Pci;

[0141] 4.4 Calculate the depth Gridid ​​of point Pi after expansion, then Gridid ​​= the modulus of vector Vi;

[0142] 4.5 Calculate the plane coordinates Gridiy after the point Pi is expanded;

[0143] 4.5.1 According to the index of point Pci0 and point Pci1 in the center line point set DC, take out the mileage Mileageci0 of point Pci0 and the mileage Mileageci1 of point Pci1 from the mileage set DCMileage respectively;

[0144] 4.5.2 Calculate the length from point Pci0 to point Pci1 as Length, and the length from point Pci to point Pci0 as Length0;

[0145] 4.5.3Gridix=(Mileageci1-Mileageci0)*Length0 / Length;

[0146] 4.6 Calculate the plane coordinates Gridiy after the point Pi is expanded;

[0147] 4.6.1 Calculate the vector Vi from the projection point Pci of point Pi to Pi on the midline, then Vi is the point Pci minus the point Pi, that is, Vi = Pi – Pci;

[0148] 4.6.2 Calculate Gridiy by the arccosine function acos, then Gridiy = acos (z value of vector Vi, Gridid);

[0149] 4.6.3 If the x value of vector Vi is less than 0, invert Gridiy.

[0150] 4.7 Based on Gridix and Gridiy, calculate the row and column coordinates of Pi in the grid;

[0151] 4.7.1 Set the x-axis grid size to scalex, then the column of the expanded point Pi in the grid is GridixCol = Gridix / scalex;

[0152] 4.7.2 Set the x-axis grid size to scaley, then the column and row of the expanded point Pi in the grid GridixRow = Gridiy / scaley;

[0153] 4.8 Store the expanded Pi depth in the two-dimensional array GridDepth, that is, GridDepth[GridixRow][GridixCol] = Gridid;

[0154] 5. Comparison of multiple point clouds

[0155] During the monitoring process, the same tunnel will be scanned multiple times, point cloud data from different periods will be collected, and the differences in point clouds from different periods will be analyzed to determine whether the tunnel has deformed. Generally, we use the first collected point cloud data as the reference point cloud, and the subsequent collected point clouds as the point clouds to be tested and the reference point clouds for comparison. If there is any construction or other changes to the tunnel, the first point cloud collected after the update needs to be updated as the reference point cloud.

[0156] 5.1 According to the above steps, expand the reference point cloud CloudTarget to obtain the two-dimensional grid depth array GridDepthTarget;

[0157] 5.2 According to the above steps, expand the point cloud Cloudn to obtain the two-dimensional grid depth array GridDepthn;

[0158] 5.3 Traverse the rows and columns of the two-dimensional grid depth array, subtract the depth value stored in GridDepthTarget from the depth value stored in the corresponding row and column, and obtain the depth difference two-dimensional array GridDepthDiffn;

[0159] 5.4 Repeat steps 5.2 and 5.3, and process the point clouds Cloudn1…Cloudnm of other periods in this way to obtain the depth difference two-dimensional array GridDepthDiffn1…GridDepthDiffnm;

[0160] 5.5 Create a color band, set less than 0.01m as green, 0.01-0.02m as yellow, and greater than 0.02m as red;

[0161] 5.6 Traverse each grid of the two-dimensional grid depth array GridDepthn, take the corresponding color through the color band according to the depth difference stored in each grid, and output the corresponding change heat map Heatmapn;

[0162] 5.7 Repeat step 5.6, process the depth difference two-dimensional arrays GridDepthDiffn1…GridDepthDiffnm of other periods in turn, and obtain the corresponding change heat maps Heatmapn1…Heatmapnm;

[0163] 6. Identification and marking of deformation areas

[0164] During the deformation monitoring process, there may be noise points that are not completely filtered out in some places that affect the judgment. In order to make the inspection more accurate, we need to combine the depth difference and the range of the difference to make the judgment, that is, there are depth deformation threshold DepthDiffThreshold and deformation area threshold DiffAreaThreshold. Only the area with a deformation area larger than DiffAreaThreshold will be marked as an undeformed dangerous area to remind manual review.

[0165] 6.1 Traverse the depth in GridDepthDiffn and compare it with the depth deformation threshold DepthDiffThreshold. If it exceeds, it means that the area where the grid is located has been deformed and marked;

[0166] 6.2 Expand the deformed grid outwards, and if the adjacent grids are also deformed grids, connect them together;

[0167] 6.3 Count the number of connected areas of the deformation grid. If it is greater than the deformation area threshold DiffAreaThreshold, it means that the area has been deformed and is marked as a dangerous area. If there is a dangerous area, the alarm system will be used to remind relevant staff to review and confirm.

[0168] 6.4 Repeat the above steps to process the multi-period depth difference two-dimensional array GridDepthDiffn1…GridDepthDiffnm and mark the corresponding danger areas.

[0169] 7. Result analysis and subsequent processing

[0170] The monitoring system provides detailed deformation reports and trend analysis based on the marking and statistical analysis of deformation areas.

[0171] 7.1 If there are multiple periods of data, the change heat maps Heatmapn1…Heatmapnm are played in sequence, each picture is a frame, and the video storage is generated to show the change trend of tunnel deformation.

[0172] 7.2 Analyze the change trend, analyze the data of the area where there is a change, but it has not reached the deformation area threshold DiffAreaThreshold, and connect with the corresponding position of the earlier data to determine whether the deformation area has a trend of increasing. If so, it is a potential deformation area, and the relevant staff will be reminded to conduct on-site inspections through email messages to promptly discover potential risks.

[0173] Example 3

[0174] This embodiment also provides a deformation monitoring system based on shield tunnel point cloud, which is used to implement the method of embodiment 1 or 2, such as Figure 3 As shown, including:

[0175] The data acquisition module scans the tunnel multiple times to obtain the reference tunnel point cloud data and the measured tunnel point cloud data at different periods;

[0176] A denoising module performs denoising on the reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods, thereby obtaining the denoised reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods;

[0177] The centerline expansion module expands the de-noised reference tunnel point cloud data and the measured tunnel point cloud data at different periods according to the designed centerline of the tunnel to obtain the corresponding reference two-dimensional grid depth array and the measured two-dimensional grid depth array at different periods;

[0178] The difference comparison module compares the depth values ​​in the corresponding reference two-dimensional grid depth array with the depth values ​​in the two-dimensional grid depth array to be tested at different periods to obtain a corresponding depth difference two-dimensional array;

[0179] The deformation determination module sets a depth deformation threshold and a deformation area threshold, compares the depth value of the corresponding depth difference two-dimensional array with the depth deformation threshold and the deformation area threshold, and determines the deformation of the tunnel based on the comparison result.

[0180] This embodiment also provides an electronic device, including:

[0181] one or more processors;

[0182] A memory for storing one or more programs;

[0183] When the one or more programs are executed by the one or more processors, the processors are enabled to implement the above method.

[0184] This embodiment also provides a computer-readable storage medium containing a computer program, on which a computer program is stored. When the program is executed by one or more processors, the above method is implemented.

[0185] The same or similar reference numerals correspond to the same or similar components;

[0186] The terms used in the drawings to describe positional relationships are only used for illustrative purposes and should not be construed as limiting this patent;

[0187] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A deformation monitoring method based on shield tunnel point cloud, characterized in that: include: Scan the tunnel multiple times to obtain reference tunnel point cloud data and test tunnel point cloud data at different times; De-noising the reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods to obtain the de-noised reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods; Based on the denoised reference tunnel point cloud data and the measured tunnel point cloud data at different periods, the data are expanded according to the designed center line of the tunnel to obtain the corresponding reference two-dimensional grid depth array and the measured two-dimensional grid depth array at different periods; Compare the depth values ​​in the corresponding reference two-dimensional grid depth array with the depth values ​​in the two-dimensional grid depth array to be tested at different periods to obtain a corresponding depth difference two-dimensional array; A depth deformation threshold and a deformation area threshold are set, and the depth value of the corresponding depth difference two-dimensional array is compared with the depth deformation threshold and the deformation area threshold, and the deformation of the tunnel is determined based on the comparison result.

2. The deformation monitoring method based on shield tunnel point cloud according to claim 1 is characterized in that: Scan the tunnel multiple times to obtain reference tunnel point cloud data and test tunnel point cloud data at different times, including: Use a total station to establish control points on the tunnel to be measured, and establish a coordinate system consistent with the target coordinate system; A three-dimensional laser instrument is set up, and three prism target balls are arranged at predetermined positions from the laser instrument, wherein the first target ball is located in the direction of tunnel advancement, the second target ball is located in the opposite direction of tunnel advancement, and the third target ball is located between the two and close to the three-dimensional laser instrument; Start the 3D laser instrument to collect raw laser data, and use the total station to collect the coordinates of the target sphere prism until the data of the entire tunnel to be measured is collected; Perform data processing, convert the original laser data into point cloud data, find the position of the target ball, intercept the point cloud of the target ball, identify and fit the point cloud coordinates of the center of all target balls; Record the point cloud coordinates of the center of the target sphere in each point cloud and the corresponding coordinates of the center of the target sphere prism collected by the total station as the same-name point pairs in different coordinate systems; Using the three pairs of points with the same name in each point cloud data, the transformation matrix from each point cloud to the target coordinate system is calculated through singular value decomposition, and all point clouds are converted to the target coordinate system to obtain the reference tunnel point cloud data and the point cloud data of the tunnel to be tested at different periods.

3. The deformation monitoring method based on shield tunnel point cloud according to claim 1 is characterized in that: The reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods are denoised to obtain the denoised reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods, including: A nearest neighbor search tree is established to process the reference tunnel point cloud data to be filtered and the tunnel point cloud data to be tested at different periods; Traversing the point cloud data, searching for all points within a preset radius through the nearest neighbor search tree, and being able to adjust the filter sensitivity through an external interface setting; For each point set within a preset radius, calculate the mean of all points; Calculate the standard deviation of the distance from each point in the point set to the mean, which is used to indicate the degree of dispersion of the point distribution; If the distance from the point to the mean is greater than the preset multiple of the standard deviation, the point is marked as a deleted point; After the traversal is completed, all marked deleted points are removed to obtain the denoised reference tunnel point cloud data and the measured tunnel point cloud data at different periods.

4. The deformation monitoring method based on shield tunnel point cloud according to claim 1 is characterized in that: Based on the denoised reference tunnel point cloud data and the measured tunnel point cloud data at different periods, the corresponding reference two-dimensional grid depth array and the measured two-dimensional grid depth array at different periods are obtained according to the designed center line of the tunnel, including: Input the corresponding design parameters according to the design drawings to obtain the design center line; Discretize the designed centerline at specified intervals to obtain a discrete centerline point set and a corresponding mileage set; Traverse the denoised reference tunnel point cloud data and calculate the projection points of the two points closest to the midline point set in the point cloud data on the designed midline; Determine the first plane coordinates of the points of the corresponding point cloud data after expansion according to the indexes of the two points closest to the midline point set in the point cloud data and the mileage data in the corresponding mileage set, Calculate the vector from the projection point of a point in the reference tunnel point cloud data on the design centerline to the point, where the modulus of the vector is the depth value of the point after expansion; According to the vector from the projection point on the design centerline to the point in the reference tunnel point cloud data, the second plane coordinates of the point corresponding to the point cloud data after expansion are calculated by the arc cosine function; Constructing a corresponding reference two-dimensional grid depth array according to the first plane coordinate and the second plane coordinate, and storing the depth values ​​of the points in the reference tunnel point cloud data after expansion in the reference two-dimensional grid depth array; According to the tunnel point cloud data to be tested at different periods, the two-dimensional grid depth arrays to be tested at different periods are constructed in accordance with the above steps, and the depth values ​​of the points in the tunnel point cloud data to be tested at different periods after expansion are stored in the two-dimensional grid depth arrays to be tested at different periods.

5. The deformation monitoring method based on shield tunnel point cloud according to claim 1 is characterized in that: A depth deformation threshold and a deformation area threshold are set, and the depth value of the corresponding depth difference two-dimensional array is compared with the depth deformation threshold and the deformation area threshold, and the deformation of the tunnel is determined based on the comparison result, including: Traverse each depth value in the depth difference two-dimensional array and compare it with the set depth deformation threshold. If the depth value exceeds the depth deformation threshold, mark the area where the grid is located as a deformation area; Expand the grid marked as deformed outwards. If the adjacent grid is also a deformed grid, connect the current deformed grid. Count the number of connected areas of the deformed grid. If the number of connected areas is greater than the preset deformation area threshold, the area is marked as a dangerous area. Process the depth difference two-dimensional arrays corresponding to all periods, and mark the dangerous areas corresponding to all depth difference two-dimensional arrays.

6. The deformation monitoring method based on shield tunnel point cloud according to claim 1 is characterized in that: Also includes: Comparing the depth values ​​in the corresponding reference two-dimensional grid depth array with the depth values ​​in the two-dimensional grid depth array to be tested at different periods, to obtain a number of corresponding depth difference two-dimensional arrays; Create a color band, traverse all the depth difference two-dimensional arrays, match the corresponding color of the color band according to the depth value in each depth difference two-dimensional array, and obtain the corresponding thermal change map.

7. The deformation monitoring method based on shield tunnel point cloud according to claim 6 is characterized in that: Also includes: Generate thermal change videos from thermal change diagrams in sequence and store them; The deformation trend is analyzed based on the thermal change video. If there is a deformation area and the deformation area has not reached the deformation area threshold, it is judged based on historical data whether the deformation area has a trend of increasing. When the deformation area has a trend of increasing, it is marked as a potential deformation area, and the staff is notified through smart devices to conduct on-site inspections.

8. A deformation monitoring system based on shield tunnel point cloud, characterized in that: The method for implementing any one of claims 1 to 7 comprises: The data acquisition module scans the tunnel multiple times to obtain the reference tunnel point cloud data and the measured tunnel point cloud data at different periods; A denoising module performs denoising on the reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods, thereby obtaining the denoised reference tunnel point cloud data and the tunnel point cloud data to be tested at different periods; The centerline expansion module expands the de-noised reference tunnel point cloud data and the measured tunnel point cloud data at different periods according to the designed centerline of the tunnel to obtain the corresponding reference two-dimensional grid depth array and the measured two-dimensional grid depth array at different periods; The difference comparison module compares the depth values ​​in the corresponding reference two-dimensional grid depth array with the depth values ​​in the two-dimensional grid depth array to be tested at different periods to obtain a corresponding depth difference two-dimensional array; The deformation determination module sets a depth deformation threshold and a deformation area threshold, compares the depth value of the corresponding depth difference two-dimensional array with the depth deformation threshold and the deformation area threshold, and determines the deformation of the tunnel based on the comparison result.

9. An electronic device, characterized in that: include: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the processors are caused to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium containing a computer program, wherein the computer program is stored thereon, characterized in that: When the program is executed by one or more processors, the method according to any one of claims 1 to 7 is implemented.

Images