Tunnel lining deformation detection method and device based on three-dimensional point cloud

Abstract

This application relates to the field of tunnel inspection technology, providing a method and apparatus for detecting tunnel lining deformation based on 3D point clouds. The method includes: acquiring the actual mileage marker location matching the cross-sectional contour data in the current target data; correcting the actual mileage marker location; acquiring the historical cross-sectional contour dataset corresponding to the corrected location in the historical target data; obtaining the local deformation and detection time difference of the tunnel lining at different locations based on the differences between the cross-sectional contour data and the historical cross-sectional contour data in the attitude-fine-tuned dataset; and stitching together multiple deformation regions of the tunnel lining to obtain the deformation detection result information of the tunnel lining. This application, through multiple registrations of the cross-sectional contour data and the actual mileage marker location, can accurately capture the deformation region of the tunnel lining. Using high-density 3D point cloud data of the tunnel lining, high-density detection results can be obtained, improving the accuracy of tunnel lining deformation detection.

Description

Technical Field

[0001] This application relates to the field of tunnel inspection technology, specifically to a method and device for detecting tunnel lining deformation based on three-dimensional point clouds. Background Technology

[0002] Tunnel operation safety is of paramount importance in tunnel maintenance. As the protective layer of the tunnel structure, the tunnel lining is subjected to external forces such as groundwater pressure and earthquakes. Over time, due to factors such as geological conditions and construction techniques, tunnel linings may experience cracks and deformations. If these cracks and deformations are not detected and addressed in a timely manner, they may have a serious impact on tunnel safety.

[0003] Existing rapid tunnel inspection methods often only focus on the surface damage of the tunnel, while the deformation of the tunnel lining is usually monitored by inclinometers, displacement gauges, strain gauges, crack gauges, etc. This monitoring method is usually a discrete section monitoring method, which cannot obtain high-density detection results, thus affecting the accuracy of tunnel lining deformation detection. Summary of the Invention

[0004] This application provides a method and apparatus for detecting tunnel lining deformation based on three-dimensional point clouds, which solves the technical problem that existing rapid tunnel inspections often only focus on the surface damage of the tunnel, while tunnel lining deformation detection methods cannot obtain high-density detection results, thus affecting the accuracy of tunnel lining deformation detection.

[0005] In a first aspect, embodiments of this application provide a method for detecting tunnel lining deformation based on three-dimensional point clouds, comprising: acquiring a first actual mileage station position that matches any cross-sectional contour data in the current target data; wherein the current target data is data after preprocessing the high-density three-dimensional point cloud data of the tunnel lining measured at the current moment; The first historical cross-sectional profile dataset is obtained from the historical target data, which is the first historical cross-sectional profile data of the area near the first actual mileage station. The historical target data is the data after preprocessing the high-density three-dimensional point cloud data of the tunnel lining measured at historical time. The first historical cross-sectional profile dataset is used to correct the station of any cross-sectional profile data to obtain the second actual mileage station. Obtain the second historical cross-sectional profile dataset corresponding to the second actual mileage station position in the historical target data; perform attitude fine-tuning on each historical cross-sectional profile data in the second historical cross-sectional profile dataset to obtain the third historical cross-sectional profile dataset. Based on the differences between any cross-sectional profile data and the historical cross-sectional profile data in the third historical cross-sectional profile dataset, the local deformation and detection time difference of any cross-sectional profile data at different positions of the tunnel lining are obtained; the local deformation corresponding to the same detection time difference are spliced ​​together to obtain multiple first deformation regions of the tunnel lining; based on the multiple first deformation regions, the deformation detection result information of the tunnel lining is obtained.

[0006] In one embodiment, obtaining the deformation detection result information of the tunnel lining based on the plurality of first deformation regions includes: The plurality of first deformation regions are extended and denoised to obtain a plurality of second deformation regions; The multiple second deformation regions are classified according to the location of the disease to obtain multiple categories of second deformation regions; For second deformation regions that have overlapping deformation areas and belong to the same category, sort them according to deformation time to obtain multiple sequences of second deformation regions; Based on the deformation evolution law model, abnormal second deformation regions in the multiple second deformation region sequences are removed to obtain multiple third deformation region sequences; Based on the multiple third deformation region sequences, the deformation detection result information of the tunnel lining is obtained; the deformation detection result information includes the position, area, and radial deformation rate of any point in each third deformation region in the third deformation region sequence.

[0007] In one embodiment, obtaining the first actual mileage marker location that matches any cross-sectional profile data in the current target data includes: Using the mapping relationship between any cross-sectional profile data and the measured mileage at the current time, and the mapping relationship between the measured mileage at the current time and the actual mileage of the tunnel, the first actual mileage station position matching any cross-sectional profile data in the current target data is obtained; or

[0008] By utilizing the mapping relationship between any cross-sectional profile data and the current time, and the mapping relationship between the current time and the actual mileage of the tunnel, the first actual mileage station position that matches any cross-sectional profile data in the current target data is obtained.

[0009] In one embodiment, the first historical cross-sectional profile dataset of the area near the first actual mileage marker location in the acquired historical target data includes: Obtain historical cross-sectional contour data within a first preset distance range from the first actual mileage marker location from the historical target data to obtain the first historical cross-sectional contour dataset.

[0010] In one embodiment, the step of using the first historical cross-sectional profile dataset to perform station correction on any of the cross-sectional profile data to obtain the second actual mileage station location includes: Extract historical feature point data of each historical cross-section contour from the first historical cross-section contour dataset, and current feature point data of any cross-section contour data. Calculate the similarity between the current feature point data and each historical feature point data, and determine the historical cross-sectional contour data to which the historical feature point data with the maximum similarity value belongs as the target cross-sectional contour data; The actual mileage marker location matched with the target cross-sectional profile data is determined as the second actual mileage marker location.

[0011] In one embodiment, obtaining the second historical cross-sectional profile dataset corresponding to the second actual mileage marker location in the historical target data includes: In any historical period of the historical target data, the cross-sectional profile data that is closest to the second actual mileage marker is obtained, and the closest cross-sectional profile data is used as the second historical cross-sectional profile data; all the second historical cross-sectional profile data in the historical target data are combined into a dataset to obtain the second historical cross-sectional profile dataset.

[0012] In one embodiment, the step of fine-tuning the attitude of each historical cross-section contour data in the second historical cross-section contour dataset to obtain a third historical cross-section contour dataset includes: Extract historical feature point data of each historical cross-section contour from the second historical cross-section contour dataset, as well as current feature point data of any cross-section contour data; Based on the positional distribution of the current feature point data, the posture of the historical feature point data of each historical cross-section contour in the second historical cross-section contour dataset is fine-tuned to obtain the third historical cross-section contour dataset.

[0013] In one embodiment, extending and denoising the plurality of first deformation regions to obtain a plurality of second deformation regions includes: The plurality of first deformation regions are sorted in descending order of area, and the plurality of first deformation regions with the highest sorted area are selected to obtain a plurality of seed regions; Multiple extended regions are obtained by extending outward from the edges of various sub-regions with a preset extension range; Among the multiple extended regions, the smaller areas are removed, and the remaining areas are identified as the second deformation regions.

[0014] In one embodiment, the target data is determined based on the following method: Using the measured attitude data, the three-dimensional point cloud data of the tunnel lining is compensated for to obtain the compensated three-dimensional point cloud data. By utilizing the continuity of the cross-sectional contour data in the compensated 3D point cloud data, abnormal abrupt points in the compensated 3D point cloud data are marked to obtain marked 3D point cloud data. Based on the chronological order of measurement times of each data point in the marked 3D point cloud data, the data points are stitched together to obtain the stitched 3D point cloud data. Using normal data points within a second preset distance range from the edge of the abnormal region to which the abnormal marker point belongs in the stitched 3D point cloud data, the abnormal marker point is interpolated and replaced to obtain the target data.

[0015] Secondly, embodiments of this application provide a tunnel lining deformation detection device based on three-dimensional point clouds, comprising: The first actual mileage station location acquisition module is used to: acquire the first actual mileage station location that matches any cross-sectional contour data in the current target data; the current target data is the data after preprocessing the high-density three-dimensional point cloud data of the tunnel lining measured at the current moment. The first historical cross-section contour dataset acquisition module is used to: acquire the first historical cross-section contour dataset of the area near the first actual mileage station in the historical target data; the historical target data is the data after preprocessing the high-density three-dimensional point cloud data of the tunnel lining measured at historical times; The second actual mileage station location acquisition module is used to: use the first historical cross-sectional profile dataset to perform station correction on any of the cross-sectional profile data to obtain the second actual mileage station location. The second historical cross-section contour dataset acquisition module is used to: acquire the second historical cross-section contour dataset corresponding to the second actual mileage station position in the historical target data. The third historical cross-section contour dataset acquisition module is used to: fine-tune the attitude of each historical cross-section contour data in the second historical cross-section contour dataset to obtain the third historical cross-section contour dataset. The cross-sectional profile data comparison module is used to: obtain the local deformation and detection time difference of any cross-sectional profile data at different positions of the tunnel lining based on the difference between any cross-sectional profile data and each historical cross-sectional profile data in the third historical cross-sectional profile dataset; The first deformation region generation module is used to: splice together the local deformations corresponding to the same detection time difference to obtain multiple first deformation regions of the tunnel lining; The tunnel lining deformation information extraction module is used to obtain the deformation detection result information of the tunnel lining based on the plurality of first deformation regions.

[0016] The tunnel lining deformation detection method and apparatus based on 3D point cloud provided in this application acquires a first actual mileage station position that matches any cross-sectional contour data in the current target data, acquires a first historical cross-sectional contour dataset of the area near the first actual mileage station position in the historical target data, performs station correction on the cross-sectional contour data using the first historical cross-sectional contour dataset to obtain a second actual mileage station position, acquires a second historical cross-sectional contour dataset corresponding to the second actual mileage station position in the historical target data, performs attitude fine-tuning on each historical cross-sectional contour data in the second historical cross-sectional contour dataset to obtain a third historical cross-sectional contour dataset, obtains the local deformation and detection time difference of the cross-sectional contour data at different positions of the tunnel lining based on the difference between the cross-sectional contour data and each historical cross-sectional contour data in the third historical cross-sectional contour dataset, splices the local deformation corresponding to the same detection time difference to obtain multiple first deformation regions of the tunnel lining, and obtains the deformation detection result information of the tunnel lining based on the multiple first deformation regions. The current target data is the preprocessed high-density 3D point cloud data of the tunnel lining measured at the current time, while the historical target data is the preprocessed high-density 3D point cloud data of the tunnel lining measured at a historical time. On the one hand, by repeatedly registering the cross-sectional contour data in the tunnel lining 3D point cloud data with the actual mileage station position, multiple accurate locations of tunnel lining deformation can be obtained. Then, by stitching together the local deformation of the accurate locations corresponding to the same detection time difference, the area of ​​tunnel lining deformation can be accurately captured, improving the accuracy of tunnel lining deformation detection. On the other hand, the high-density 3D point cloud data of the tunnel lining can obtain high-density detection results, further improving the accuracy of tunnel lining deformation detection. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is one of the flowcharts of the tunnel lining deformation detection method based on three-dimensional point cloud provided in the embodiments of this application; Figure 2 This is the second flowchart of the tunnel lining deformation detection method based on three-dimensional point cloud provided in the embodiments of this application; Figure 3 This is the third flowchart of the tunnel lining deformation detection method based on three-dimensional point cloud provided in the embodiments of this application; Figure 4 This is the fourth flowchart of the tunnel lining deformation detection method based on three-dimensional point cloud provided in the embodiments of this application; Figure 5 This is the fifth flowchart of the tunnel lining deformation detection method based on three-dimensional point cloud provided in the embodiments of this application; Figure 6 This is a schematic diagram of the structure of the tunnel lining deformation detection device based on three-dimensional point cloud provided in the embodiments of this application; Figure 7 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] It should be noted that in the description of the embodiments of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. The terms "upper," "lower," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Unless otherwise expressly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly, for example, they can be fixed connections, detachable connections, or integral connections; they can be mechanical connections or electrical connections; they can be direct connections or indirect connections through an intermediate medium; and they can be internal connections between two elements. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0021] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, a first object can be one or more. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects have an "or" relationship.

[0022] Figure 1 This is one of the flowcharts illustrating the tunnel lining deformation detection method based on three-dimensional point clouds provided in this application embodiment. (Refer to...) Figure 1 This application provides a tunnel lining deformation detection method based on three-dimensional point cloud, which may include: 101. Obtaining the first actual mileage station position that matches any cross-sectional contour data in the current target data; The current target data is the preprocessed data of the high-density three-dimensional point cloud data of the tunnel lining measured at the current time; 102. Obtain the first historical cross-sectional profile dataset of the area near the first actual mileage station in the historical target data; the historical target data is the preprocessed data of the high-density three-dimensional point cloud data of the tunnel lining measured at a historical time; 103. Use the first historical cross-sectional profile dataset to perform station correction on the cross-sectional profile data to obtain the second actual mileage station position; 104. Obtain the second historical cross-sectional profile dataset corresponding to the second actual mileage station position in the historical target data; 105. Fine-tune the attitude of each historical cross-section contour data in the second historical cross-section contour dataset to obtain the third historical cross-section contour dataset. 106. Based on the differences between the cross-sectional profile data and the historical cross-sectional profile data in the third historical cross-sectional profile dataset, the local deformation and detection time difference of the cross-sectional profile data at different locations of the tunnel lining are obtained. 107. By splicing together the local deformations corresponding to the same detection time difference, multiple first deformation regions of the tunnel lining are obtained; 108. Based on multiple first deformation regions, obtain the deformation detection results information of the tunnel lining.

[0023] In step 103, based on the matching of corresponding feature points and the least squares estimation method, the similarity between the current cross-sectional profile and each profile in the first historical cross-sectional profile dataset can be used to correct the station number corresponding to the cross-sectional profile data, so as to obtain the location of the second actual mileage station number.

[0024] In step 106, the local deformation of the cross-sectional profile data at different locations in the tunnel lining can be obtained based on the relationship between the obtained differences and the preset deformation thresholds for the arch crown, arch waist, and arch bottom.

[0025] In each inspection, high-density three-dimensional point cloud data of the tunnel lining can be obtained using a three-dimensional measurement sensor, the measurement attitude of the three-dimensional measurement sensor can be obtained using an inertial navigation system, the detection position information can be obtained using a positioning system, and the detection time of the three-dimensional point cloud data can be recorded; wherein, the lateral measurement point spacing of the high-density three-dimensional point cloud data is less than 10mm, and the longitudinal cross-sectional spacing is less than 100mm.

[0026] Furthermore, the 3D measurement sensor, inertial navigation system, and positioning system are installed on the same vehicle platform. Through the control system, 3D point cloud data, attitude measurement, position information detection, and detection time can be acquired synchronously. The 3D measurement sensor can be one or more of a 3D LiDAR device or a line-scan 3D measurement sensor; no limitation is made here.

[0027] The tunnel lining deformation detection method based on 3D point cloud provided in this embodiment obtains the first actual mileage station position that matches any cross-sectional contour data in the current target data, obtains the first historical cross-sectional contour dataset of the area near the first actual mileage station position in the historical target data, performs station correction on the cross-sectional contour data using the first historical cross-sectional contour dataset to obtain the second actual mileage station position, obtains the second historical cross-sectional contour dataset corresponding to the second actual mileage station position in the historical target data, performs attitude fine-tuning on each historical cross-sectional contour data in the second historical cross-sectional contour dataset to obtain the third historical cross-sectional contour dataset, obtains the local deformation and detection time difference of the cross-sectional contour data at different positions of the tunnel lining based on the difference between the cross-sectional contour data and each historical cross-sectional contour data in the third historical cross-sectional contour dataset, splices the local deformation corresponding to the same detection time difference to obtain multiple first deformation regions of the tunnel lining, and obtains the deformation detection result information of the tunnel lining based on the multiple first deformation regions. The current target data is the preprocessed high-density 3D point cloud data of the tunnel lining measured at the current time, while the historical target data is the preprocessed high-density 3D point cloud data of the tunnel lining measured at a historical time. On the one hand, by repeatedly registering the cross-sectional contour data in the tunnel lining 3D point cloud data with the actual mileage station position, multiple accurate locations of tunnel lining deformation can be obtained. Then, by stitching together the local deformation of the accurate locations corresponding to the same detection time difference, the area of ​​tunnel lining deformation can be accurately captured, improving the accuracy of tunnel lining deformation detection. On the other hand, the high-density 3D point cloud data of the tunnel lining can obtain high-density detection results, further improving the accuracy of tunnel lining deformation detection.

[0028] Furthermore, compared to traditional deformation detection methods that require the independent installation of monitoring equipment in each tunnel, which makes long-term maintenance of the monitoring equipment difficult and consumes a lot of manpower and resources, this embodiment installs all monitoring equipment on the same vehicle platform. By leveraging the mobility of the vehicle platform, multiple tunnels can be inspected using only one set of tunnel detection equipment. This reduces the pressure on equipment maintenance and the consumption of manpower and resources, lowers tunnel operating costs, and enables rapid detection of tunnel lining deformation under low-cost operation.

[0029] Figure 2 This is the second schematic flowchart of the tunnel lining deformation detection method based on three-dimensional point clouds provided in the embodiments of this application. (Refer to...) Figure 2 In one embodiment, obtaining deformation detection result information of the tunnel lining based on multiple first deformation regions may include: 201. Extend and denoise multiple first deformation regions to obtain multiple second deformation regions; 202. Classify multiple second deformation areas according to the location of the disease to obtain multiple categories of second deformation areas; 203. For second deformation regions that have overlapping deformation areas and belong to the same category, sort them according to deformation time to obtain multiple sequences of second deformation regions; 204. Based on the deformation evolution law model, abnormal second deformation regions are removed from multiple second deformation region sequences to obtain multiple third deformation region sequences; 205. Based on multiple third deformation region sequences, the deformation detection results of the tunnel lining are obtained.

[0030] The deformation detection results include the location, area, and radial deformation rate of any point in each of the third deformation regions in the third deformation region sequence.

[0031] In steps 202 to 204, since the deformation of the tunnel lining is usually located at different positions of the lining, and the changes in deformation at different positions are usually different in the time dimension, the deformation areas are first classified according to the deformation location, and then the deformation of each category is sorted by time and compared with the predetermined deformation evolution law scale of each position. Areas that do not conform to the deformation law are eliminated, and finally, the deformation area sequence of each category is obtained.

[0032] The scale of deformation evolution laws can be based on expert experience models or models constructed using a large amount of historical data combined with artificial intelligence.

[0033] This embodiment obtains a sequence of deformation regions that conform to the time evolution law at each location of the tunnel lining by denoising and extending the deformation region, classifying its location, comparing its evolution law, and removing abnormal regions. Then, the deformation detection results of the tunnel lining are obtained based on these deformation region sequences. The detection results eliminate the interference of abnormal regions to the greatest extent, thus further improving the accuracy.

[0034] In a real-time example, obtaining the first actual mileage marker location that matches any cross-sectional profile data in the current target data may include: By utilizing the mapping relationship between the cross-sectional profile data and the measured mileage at the current time, and the mapping relationship between the measured mileage at the current time and the actual mileage of the tunnel, the first actual mileage station position matching the cross-sectional profile data in the current target data is obtained; or

[0035] By utilizing the mapping relationship between the cross-sectional profile data and the current time, and the mapping relationship between the current time and the actual mileage of the tunnel, the first actual mileage station position that matches the cross-sectional profile data in the current target data is obtained.

[0036] The positioning system includes an encoder positioning system, and one or more of a control network sensing system and a target positioning sensing system, used to obtain the mapping relationship between the current detected mileage and the actual mileage of the tunnel. The encoder positioning system is used to obtain the current detected mileage, while one or more of the control network sensing system and the target positioning sensing system are used to obtain the actual mileage of the tunnel.

[0037] In addition, since the collection time is recorded simultaneously when collecting cross-sectional profile data and the actual mileage of the tunnel, i.e. the detection time of this detection, the cross-sectional profile data and the actual mileage of the tunnel can also be associated through the current detection time.

[0038] This embodiment can easily match the actual tunnel mileage corresponding to the cross-sectional profile data by measuring the mileage at the current time or mapping and associating the current time with the cross-sectional profile data and the actual tunnel mileage, thereby obtaining the station position corresponding to the actual tunnel mileage.

[0039] In one embodiment, obtaining a first historical cross-sectional profile dataset of the area near the first actual mileage marker in historical target data may include: The first historical cross-sectional profile dataset is obtained by acquiring historical cross-sectional profile data within a first preset distance range from the first actual mileage marker location in the historical target data.

[0040] In this embodiment, by selecting historical cross-sectional profile data close to the first actual mileage marker as the first historical cross-sectional profile dataset, it is possible to obtain a historical cross-sectional profile dataset that is as close as possible to the current cross-sectional profile data.

[0041] Figure 3 This is the third flowchart illustrating the tunnel lining deformation detection method based on three-dimensional point clouds provided in this application. (Refer to...) Figure 3 In one embodiment, using a first historical cross-sectional profile dataset to perform stationing correction on any cross-sectional profile data to obtain the second actual mileage stationing location may include: 301. Extract the historical feature point data of each historical cross-section contour in the first historical cross-section contour dataset, as well as the current feature point data of the cross-section contour data; 302. Calculate the similarity between the current feature point data and each historical feature point data, and determine the historical cross-sectional contour data to which the historical feature point data with the maximum similarity value belongs as the target cross-sectional contour data; 303. The actual mileage station position matched with the target cross-sectional profile data is determined as the second actual mileage station position.

[0042] In step 301, both historical feature points and current feature points can include corner points, points with smaller curvature, points with larger curvature, etc., and there is no limitation here.

[0043] In step 302, any similarity calculation method can be used to calculate the similarity between the current feature point data and each historical feature point data. No limitation is made here. In this embodiment, the reciprocal of the matching error obtained by the least squares method can be used as the similarity.

[0044] This embodiment uses feature point comparison to filter out the historical cross-sectional contour data that is most similar to the current cross-sectional contour data, and determines the actual mileage station position that matches the historical cross-sectional contour data as the second actual mileage station position. That is, the station position of the current cross-sectional data is corrected by using the station position of the historical similar cross-sectional data, and by fully combining historical data experience, the accurate position of the current cross-sectional contour data is obtained.

[0045] In one embodiment, obtaining the second historical cross-sectional profile dataset corresponding to the second actual mileage marker location in the historical target data may include: In any historical period of the historical target data, the cross-sectional profile data that is closest to the second actual mileage station is obtained, and the closest cross-sectional profile data is used as the second historical cross-sectional profile data. All the second historical cross-sectional profile data in the historical target data are combined into a dataset to obtain the second historical cross-sectional profile dataset.

[0046] In this embodiment, each detection includes only one detection period, which includes multiple detection times. By acquiring the cross-sectional contour data closest to the second actual mileage marker within all historical detection periods from the historical target data, it is possible to obtain as many historical cross-sectional contour datasets as possible that are close to the current cross-sectional contour data after position correction.

[0047] In one embodiment, the pose fine-tuning of each historical cross-section contour data in the second historical cross-section contour dataset to obtain the third historical cross-section contour dataset may include: Extract historical feature point data of each historical cross-section contour from the second historical cross-section contour dataset, as well as current feature point data of the cross-section contour data. Based on the positional distribution of the current feature point data, fine-tune the posture of the historical feature point data of each historical cross-section contour in the second historical cross-section contour dataset to obtain the third historical cross-section contour dataset.

[0048] The historical second cross-section contour can be rotated to adjust the positions of historical feature points of each historical cross-section contour in the historical second cross-section contour dataset to be consistent with the positions of the current feature points.

[0049] In this embodiment, based on the positional distribution of the current feature point data, the orientation of the historical feature point data of each historical cross-section contour in the second historical cross-section contour dataset is finely adjusted so that the positions of the data points of each historical cross-section contour in the obtained third historical cross-section contour dataset are as consistent as possible with the positions of the current cross-section contour, which facilitates subsequent comparison of data at the same positions and improves the reliability of the comparison.

[0050] Figure 4 This is the fourth flowchart illustrating the tunnel lining deformation detection method based on three-dimensional point clouds provided in this application. (Refer to...) Figure 4 In one embodiment, extending and denoising multiple first deformation regions to obtain multiple second deformation regions may include: 401. Sort the multiple first deformation regions in descending order of area, and select the first deformation regions at the top of the sort to obtain multiple seed regions; 402. Extend outward from the edges of various sub-regions with a preset extension range to obtain multiple extended regions; 403. In the multiple extended regions, the smaller regions are removed, and the remaining regions are identified as the second deformation regions.

[0051] This embodiment improves the extension efficiency by extending the larger deformation region outward. Then, based on the preset area threshold Ta, regions with areas smaller than Ta are removed, thus eliminating the interference of noise regions and obtaining a second deformation region that can accurately characterize multiple first deformation regions while removing interference from redundant regions.

[0052] Figure 5 This is the fifth flowchart illustrating the tunnel lining deformation detection method based on three-dimensional point clouds provided in this application. (Refer to...) Figure 5 In one embodiment, the target data can be determined based on the following method: 501. Using the measured attitude data, the three-dimensional point cloud data of the tunnel lining is measured and the attitude is compensated to obtain the compensated three-dimensional point cloud data. 502. By utilizing the continuity of cross-sectional contour data in the compensated 3D point cloud data, abnormal abrupt change points in the compensated 3D point cloud data are marked to obtain marked 3D point cloud data. 503. Based on the chronological order of measurement times of each data point in the marked 3D point cloud data, the data points are stitched together to obtain the stitched 3D point cloud data. 504. Using normal data points within a second preset distance range from the edge of the abnormal region to which the abnormal marker point belongs in the stitched 3D point cloud data, the abnormal marker point is interpolated and replaced to obtain the target data.

[0053] This embodiment preprocesses the 3D point cloud data by measuring attitude compensation, marking abnormal 3D point cloud data, stitching 3D contour data, and interpolating abnormal region data, making the obtained target data more accurate and reliable, providing a good data foundation for subsequent steps, thereby improving the accuracy of tunnel lining deformation detection results.

[0054] The tunnel lining deformation detection device based on three-dimensional point cloud provided in the embodiments of this application is described below. The tunnel lining deformation detection device based on three-dimensional point cloud described below can be referred to in correspondence with the tunnel lining deformation detection method based on three-dimensional point cloud described above.

[0055] Figure 6 This is a schematic diagram of the tunnel lining deformation detection device based on three-dimensional point clouds provided in an embodiment of this application. (Refer to...) Figure 6 This application provides a tunnel lining deformation detection device based on three-dimensional point clouds, which may include: The first actual mileage station location acquisition module 601 is used to: acquire the first actual mileage station location that matches any cross-sectional contour data in the current target data; the current target data is data after preprocessing the high-density three-dimensional point cloud data of the tunnel lining measured at the current moment. The first historical cross-section contour dataset acquisition module 602 is used to: acquire the first historical cross-section contour dataset of the area near the first actual mileage station in the historical target data; the historical target data is the data after preprocessing the high-density three-dimensional point cloud data of the tunnel lining measured at historical times; The second actual mileage station location acquisition module 603 is used to: use the first historical cross-sectional profile dataset to perform station correction on any of the cross-sectional profile data to obtain the second actual mileage station location. The second historical cross-section contour dataset acquisition module 604 is used to: acquire the second historical cross-section contour dataset corresponding to the second actual mileage station position in the historical target data. The third historical cross-section contour dataset acquisition module 605 is used to: fine-tune the attitude of each historical cross-section contour data in the second historical cross-section contour dataset to obtain the third historical cross-section contour dataset. The cross-sectional profile data comparison module 606 is used to: obtain the local deformation and detection time difference of any cross-sectional profile data at different positions of the tunnel lining based on the difference between any cross-sectional profile data and each historical cross-sectional profile data in the third historical cross-sectional profile dataset. The first deformation region generation module 607 is used to: splice together the local deformations corresponding to the same detection time difference to obtain multiple first deformation regions of the tunnel lining; The tunnel lining deformation information extraction module 608 is used to obtain the deformation detection result information of the tunnel lining based on the plurality of first deformation regions.

[0056] The tunnel lining deformation detection device based on 3D point cloud provided in this embodiment acquires the first actual mileage station position that matches any cross-sectional contour data in the current target data, acquires the first historical cross-sectional contour dataset of the area near the first actual mileage station position in the historical target data, performs station correction on the cross-sectional contour data using the first historical cross-sectional contour dataset to obtain the second actual mileage station position, acquires the second historical cross-sectional contour dataset corresponding to the second actual mileage station position in the historical target data, performs attitude fine-tuning on each historical cross-sectional contour data in the second historical cross-sectional contour dataset to obtain the third historical cross-sectional contour dataset, obtains the local deformation and detection time difference of the cross-sectional contour data at different positions of the tunnel lining based on the difference between the cross-sectional contour data and each historical cross-sectional contour data in the third historical cross-sectional contour dataset, splices the local deformation corresponding to the same detection time difference to obtain multiple first deformation regions of the tunnel lining, and obtains the deformation detection result information of the tunnel lining based on the multiple first deformation regions. The current target data is the preprocessed high-density 3D point cloud data of the tunnel lining measured at the current time, while the historical target data is the preprocessed high-density 3D point cloud data of the tunnel lining measured at a historical time. On the one hand, by repeatedly registering the cross-sectional contour data in the tunnel lining 3D point cloud data with the actual mileage station position, multiple accurate locations of tunnel lining deformation can be obtained. Then, by stitching together the local deformation of the accurate locations corresponding to the same detection time difference, the area of ​​tunnel lining deformation can be accurately captured, improving the accuracy of tunnel lining deformation detection. On the other hand, the high-density 3D point cloud data of the tunnel lining can obtain high-density detection results, further improving the accuracy of tunnel lining deformation detection.

[0057] Furthermore, compared to traditional deformation detection methods that require the independent installation of monitoring equipment in each tunnel, which makes long-term maintenance of the monitoring equipment difficult and consumes a lot of manpower and resources, this embodiment installs all monitoring equipment on the same vehicle platform. By leveraging the mobility of the vehicle platform, multiple tunnels can be inspected using only one set of tunnel detection equipment. This reduces the pressure on equipment maintenance and the consumption of manpower and resources, lowers tunnel operating costs, and enables rapid detection of tunnel lining deformation under low-cost operation.

[0058] In one embodiment, the tunnel lining deformation information extraction module 608 is specifically used for: The plurality of first deformation regions are extended and denoised to obtain a plurality of second deformation regions; The multiple second deformation regions are classified according to the location of the disease to obtain multiple categories of second deformation regions; For second deformation regions that have overlapping deformation areas and belong to the same category, sort them according to deformation time to obtain multiple sequences of second deformation regions; Based on the deformation evolution law model, abnormal second deformation regions in the multiple second deformation region sequences are removed to obtain multiple third deformation region sequences; Based on the multiple third deformation region sequences, the deformation detection result information of the tunnel lining is obtained; the deformation detection result information includes the position, area, and radial deformation rate of any point in each third deformation region in the third deformation region sequence.

[0059] In one embodiment, the first actual mileage marker location acquisition module 601 is specifically used for: Using the mapping relationship between any cross-sectional profile data and the measured mileage at the current time, and the mapping relationship between the measured mileage at the current time and the actual mileage of the tunnel, the first actual mileage station position matching any cross-sectional profile data in the current target data is obtained; or

[0060] By utilizing the mapping relationship between any cross-sectional profile data and the current time, and the mapping relationship between the current time and the actual mileage of the tunnel, the first actual mileage station position that matches any cross-sectional profile data in the current target data is obtained.

[0061] In one embodiment, the first historical cross-sectional contour dataset acquisition module 602 is specifically used for: Obtain historical cross-sectional contour data within a first preset distance range from the first actual mileage marker location from the historical target data to obtain the first historical cross-sectional contour dataset.

[0062] In one embodiment, the second actual mileage marker location acquisition module 603 is specifically used for: Extract historical feature point data of each historical cross-section contour from the first historical cross-section contour dataset, and current feature point data of any cross-section contour data. Calculate the similarity between the current feature point data and each historical feature point data, and determine the historical cross-sectional contour data to which the historical feature point data with the maximum similarity value belongs as the target cross-sectional contour data; The actual mileage marker location matched with the target cross-sectional profile data is determined as the second actual mileage marker location.

[0063] In one embodiment, the second historical cross-sectional contour dataset acquisition module 604 is specifically used for: In any historical period of the historical target data, the cross-sectional profile data that is closest to the second actual mileage marker is obtained, and the closest cross-sectional profile data is used as the second historical cross-sectional profile data; all the second historical cross-sectional profile data in the historical target data are combined into a dataset to obtain the second historical cross-sectional profile dataset.

[0064] In one embodiment, the third historical cross-sectional contour dataset acquisition module 605 is specifically used for: Extract historical feature point data of each historical cross-section contour from the second historical cross-section contour dataset, and current feature point data of any cross-section contour data.

[0065] Based on the positional distribution of the current feature point data, the posture of the historical feature point data of each historical cross-section contour in the second historical cross-section contour dataset is fine-tuned to obtain the third historical cross-section contour dataset.

[0066] In one embodiment, the tunnel lining deformation information extraction module 608 is specifically used for: The plurality of first deformation regions are sorted in descending order of area, and the plurality of first deformation regions with the highest sorted area are selected to obtain a plurality of seed regions; Multiple extended regions are obtained by extending outward from the edges of various sub-regions with a preset extension range; Among the multiple extended regions, the smaller areas are removed, and the remaining areas are identified as the second deformation regions.

[0067] In one embodiment, the target data acquisition module (not shown in the figure) is used for: Using the measured attitude data, the three-dimensional point cloud data of the tunnel lining is compensated for to obtain the compensated three-dimensional point cloud data. By utilizing the continuity of the cross-sectional contour data in the compensated 3D point cloud data, abnormal abrupt points in the compensated 3D point cloud data are marked to obtain marked 3D point cloud data. Based on the chronological order of measurement times of each data point in the marked 3D point cloud data, the data points are stitched together to obtain the stitched 3D point cloud data. Using normal data points within a second preset distance range from the edge of the abnormal region to which the abnormal marker point belongs in the stitched 3D point cloud data, the abnormal marker point is interpolated and replaced to obtain the target data.

[0068] Figure 7 An example is a schematic diagram of the structure of an electronic device, such as... Figure 7As shown, the electronic device may include a processor 710, a communication interface 720, a memory 730, and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other via the communication bus 740. The processor 710 can call a computer program in the memory 730 to execute the steps of a tunnel lining deformation detection method based on three-dimensional point clouds, such as including: Obtain the first actual mileage station position that matches any cross-sectional contour data in the current target data; the current target data is the preprocessed data of the high-density three-dimensional point cloud data of the tunnel lining measured at the current moment; The first historical cross-sectional profile dataset is obtained from the historical target data, which is the first historical cross-sectional profile data of the area near the first actual mileage station. The historical target data is the data after preprocessing the high-density three-dimensional point cloud data of the tunnel lining measured at historical time. The first historical cross-sectional profile dataset is used to correct the station of any cross-sectional profile data to obtain the second actual mileage station. Obtain the second historical cross-sectional profile dataset corresponding to the second actual mileage station position in the historical target data; perform attitude fine-tuning on each historical cross-sectional profile data in the second historical cross-sectional profile dataset to obtain the third historical cross-sectional profile dataset. Based on the difference between any cross-sectional profile data and each historical cross-sectional profile data in the third historical cross-sectional profile dataset, the local deformation and detection time difference of any cross-sectional profile data at different positions of the tunnel lining are obtained; the local deformation corresponding to the same detection time difference are spliced ​​together to obtain multiple first deformation regions of the tunnel lining. Based on the multiple first deformation regions, the deformation detection results of the tunnel lining are obtained.

[0069] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0070] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the steps of the tunnel lining deformation detection method based on three-dimensional point clouds provided in the above embodiments, such as including: Obtain the first actual mileage station position that matches any cross-sectional contour data in the current target data; the current target data is the preprocessed data of the high-density three-dimensional point cloud data of the tunnel lining measured at the current moment; The first historical cross-sectional profile dataset is obtained from the historical target data, which is the first historical cross-sectional profile data of the area near the first actual mileage station. The historical target data is the data after preprocessing the high-density three-dimensional point cloud data of the tunnel lining measured at historical time. The first historical cross-sectional profile dataset is used to correct the station of any cross-sectional profile data to obtain the second actual mileage station. Obtain the second historical cross-sectional profile dataset corresponding to the second actual mileage station position in the historical target data; perform attitude fine-tuning on each historical cross-sectional profile data in the second historical cross-sectional profile dataset to obtain the third historical cross-sectional profile dataset. Based on the difference between any cross-sectional profile data and each historical cross-sectional profile data in the third historical cross-sectional profile dataset, the local deformation and detection time difference of any cross-sectional profile data at different positions of the tunnel lining are obtained; the local deformation corresponding to the same detection time difference are spliced ​​together to obtain multiple first deformation regions of the tunnel lining. Based on the multiple first deformation regions, the deformation detection results of the tunnel lining are obtained.

[0071] On the other hand, embodiments of this application also provide a non-transitory computer-readable storage medium storing a computer program thereon, the computer program being used to cause a processor to execute the steps of the tunnel lining deformation detection method based on three-dimensional point clouds provided in the above embodiments, for example including: Obtain the first actual mileage station position that matches any cross-sectional contour data in the current target data; the current target data is the preprocessed data of the high-density three-dimensional point cloud data of the tunnel lining measured at the current moment; The first historical cross-sectional profile dataset is obtained from the historical target data, which is the first historical cross-sectional profile data of the area near the first actual mileage station. The historical target data is the data after preprocessing the high-density three-dimensional point cloud data of the tunnel lining measured at historical time. The first historical cross-sectional profile dataset is used to correct the station of any cross-sectional profile data to obtain the second actual mileage station. Obtain the second historical cross-sectional profile dataset corresponding to the second actual mileage station position in the historical target data; perform attitude fine-tuning on each historical cross-sectional profile data in the second historical cross-sectional profile dataset to obtain the third historical cross-sectional profile dataset. Based on the difference between any cross-sectional profile data and each historical cross-sectional profile data in the third historical cross-sectional profile dataset, the local deformation and detection time difference of any cross-sectional profile data at different positions of the tunnel lining are obtained; the local deformation corresponding to the same detection time difference are spliced ​​together to obtain multiple first deformation regions of the tunnel lining. Based on the multiple first deformation regions, the deformation detection results of the tunnel lining are obtained.

[0072] The non-transitory computer-readable storage medium can be any available medium or data storage device that the processor can access, including but not limited to magnetic memory (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical memory (e.g., CD, DVD, BD, HVD), and semiconductor memory (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).

[0073] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0074] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for detecting tunnel lining deformation based on three-dimensional point clouds, characterized in that, include: Obtain the first actual mileage station position that matches any cross-sectional profile data in the current target data; The current target data is the preprocessed data of the high-density three-dimensional point cloud data of the tunnel lining measured at the current moment; The first historical cross-sectional profile dataset of the area near the first actual mileage marker is obtained from the historical target data; the historical target data is the preprocessed data of high-density three-dimensional point cloud data of the tunnel lining measured at historical times. Using the first historical cross-sectional profile dataset, station correction is performed on any of the cross-sectional profile data to obtain the second actual mileage station location, including: Based on the matching of corresponding feature points and the least squares estimation method, the station number corresponding to the cross-section contour data is corrected by using the similarity between the current cross-section contour and each contour in the first historical cross-section contour dataset, so as to obtain the second actual mileage station number position. Obtain the second historical cross-sectional profile dataset corresponding to the second actual mileage station position in the historical target data; The pose of each historical cross-section contour data in the second historical cross-section contour dataset is fine-tuned to obtain the third historical cross-section contour dataset, which includes: Rotate each historical cross-section contour in the second historical cross-section contour dataset to adjust the position of the historical feature point of each historical cross-section contour in the second historical cross-section contour dataset to be consistent with the position of the current feature point, so as to obtain the third historical cross-section contour dataset. Based on the difference between any cross-sectional profile data and each historical cross-sectional profile data in the third historical cross-sectional profile dataset, the local deformation and detection time difference of any cross-sectional profile data at different locations in the tunnel lining are obtained. By splicing together the local deformations corresponding to the same detection time difference, multiple first deformation regions of the tunnel lining are obtained; Based on the multiple first deformation regions, the deformation detection results of the tunnel lining are obtained.

2. The tunnel lining deformation detection method based on three-dimensional point cloud according to claim 1, characterized in that, The step of obtaining the deformation detection result information of the tunnel lining based on the plurality of first deformation regions includes: The plurality of first deformation regions are extended and denoised to obtain a plurality of second deformation regions; The multiple second deformation regions are classified according to the location of the disease to obtain multiple categories of second deformation regions; For second deformation regions that have overlapping deformation areas and belong to the same category, sort them according to deformation time to obtain multiple sequences of second deformation regions; Based on the deformation evolution law model, abnormal second deformation regions in the multiple second deformation region sequences are removed to obtain multiple third deformation region sequences; Based on the multiple third deformation region sequences, the deformation detection result information of the tunnel lining is obtained; the deformation detection result information includes the position, area, and radial deformation rate of any point in each third deformation region in the third deformation region sequence.

3. The tunnel lining deformation detection method based on three-dimensional point cloud according to claim 1, characterized in that, The step of obtaining the first actual mileage marker location that matches any cross-sectional contour data in the current target data includes: Using the mapping relationship between any cross-sectional profile data and the measured mileage at the current time, and the mapping relationship between the measured mileage at the current time and the actual mileage of the tunnel, the first actual mileage station position matching any cross-sectional profile data in the current target data is obtained; or By utilizing the mapping relationship between any cross-sectional profile data and the current time, and the mapping relationship between the current time and the actual mileage of the tunnel, the first actual mileage station position that matches any cross-sectional profile data in the current target data is obtained.

4. The tunnel lining deformation detection method based on three-dimensional point cloud according to claim 1, characterized in that, The first historical cross-sectional contour dataset of the area near the first actual mileage marker location, which is part of the historical target data acquisition, includes: Obtain historical cross-sectional contour data within a first preset distance range from the first actual mileage marker location from the historical target data to obtain the first historical cross-sectional contour dataset.

5. The tunnel lining deformation detection method based on three-dimensional point cloud according to claim 1, characterized in that, The step of using the first historical cross-sectional profile dataset to perform station correction on any of the cross-sectional profile data to obtain the second actual mileage station location includes: Extract historical feature point data of each historical cross-section contour from the first historical cross-section contour dataset, and current feature point data of any cross-section contour data. Calculate the similarity between the current feature point data and each historical feature point data, and determine the historical cross-sectional contour data to which the historical feature point data with the maximum similarity value belongs as the target cross-sectional contour data; The actual mileage marker location matched with the target cross-sectional profile data is determined as the second actual mileage marker location.

6. The tunnel lining deformation detection method based on three-dimensional point cloud according to claim 1, characterized in that, The acquisition of the second historical cross-sectional profile dataset corresponding to the second actual mileage marker location in the historical target data includes: In any historical period of the historical target data, the cross-sectional profile data that is closest to the second actual mileage station location is obtained, and the closest cross-sectional profile data is used as the second historical cross-sectional profile data. The second historical cross-sectional contour data from the historical target data are combined into a dataset to obtain the second historical cross-sectional contour dataset.

7. The tunnel lining deformation detection method based on three-dimensional point cloud according to claim 1, characterized in that, The process of fine-tuning the pose of each historical cross-section contour data in the second historical cross-section contour dataset to obtain the third historical cross-section contour dataset includes: Extract historical feature point data of each historical cross-section contour from the second historical cross-section contour dataset, as well as current feature point data of any cross-section contour data; Based on the positional distribution of the current feature point data, the posture of the historical feature point data of each historical cross-section contour in the second historical cross-section contour dataset is fine-tuned to obtain the third historical cross-section contour dataset.

8. The tunnel lining deformation detection method based on three-dimensional point cloud according to claim 2, characterized in that, The process of extending and denoising the plurality of first deformation regions to obtain a plurality of second deformation regions includes: The plurality of first deformation regions are sorted in descending order of area, and the plurality of first deformation regions with the highest sorted area are selected to obtain a plurality of seed regions; Multiple extended regions are obtained by extending outward from the edges of various sub-regions with a preset extension range; Among the multiple extended regions, the smaller areas are removed, and the remaining areas are identified as the second deformation regions.

9. The tunnel lining deformation detection method based on three-dimensional point cloud according to claim 1, characterized in that, The target data was determined based on the following method: Using the measured attitude data, the three-dimensional point cloud data of the tunnel lining is compensated for to obtain the compensated three-dimensional point cloud data. By utilizing the continuity of the cross-sectional contour data in the compensated 3D point cloud data, abnormal abrupt points in the compensated 3D point cloud data are marked to obtain marked 3D point cloud data. Based on the chronological order of measurement times of each data point in the marked 3D point cloud data, the data points are stitched together to obtain the stitched 3D point cloud data. Using normal data points within a second preset distance range from the edge of the abnormal region to which the abnormal marker point belongs in the stitched 3D point cloud data, the abnormal marker point is interpolated and replaced to obtain the target data.

10. A tunnel lining deformation detection device based on three-dimensional point clouds, characterized in that, The method for performing tunnel lining deformation detection based on three-dimensional point clouds as described in claim 1 includes: The first actual mileage station location acquisition module is used to: acquire the first actual mileage station location that matches any cross-sectional contour data in the current target data; the current target data is the data after preprocessing the high-density three-dimensional point cloud data of the tunnel lining measured at the current moment. The first historical cross-section contour dataset acquisition module is used to: acquire the first historical cross-section contour dataset of the area near the first actual mileage station in the historical target data; the historical target data is the data after preprocessing the high-density three-dimensional point cloud data of the tunnel lining measured at historical times; The second actual mileage station location acquisition module is used to: use the first historical cross-sectional profile dataset to perform station correction on any of the cross-sectional profile data to obtain the second actual mileage station location. The second historical cross-section contour dataset acquisition module is used to: acquire the second historical cross-section contour dataset corresponding to the second actual mileage station position in the historical target data. The third historical cross-section contour dataset acquisition module is used to: fine-tune the attitude of each historical cross-section contour data in the second historical cross-section contour dataset to obtain the third historical cross-section contour dataset. The cross-sectional profile data comparison module is used to: obtain the local deformation and detection time difference of any cross-sectional profile data at different positions of the tunnel lining based on the difference between any cross-sectional profile data and each historical cross-sectional profile data in the third historical cross-sectional profile dataset; The first deformation region generation module is used to: splice together the local deformations corresponding to the same detection time difference to obtain multiple first deformation regions of the tunnel lining; The tunnel lining deformation information extraction module is used to obtain the deformation detection result information of the tunnel lining based on the plurality of first deformation regions.

Images