Tunneling rebar ranging method, system, electronic device, and storage medium

Abstract

The application provides a tunnel reinforcement mesh ranging method and system, electronic equipment and a storage medium. The method comprises preprocessing tunnel initial point cloud data to obtain down-sampling point cloud data, and filtering non-target point cloud data through a preset algorithm to obtain target point cloud data of the reinforcement mesh to be measured; combining section design data, using an extended coordinate reflection algorithm to calculate the service feature information between each point in the target point cloud data and its corresponding design point in its section, using a KMean clustering analysis method to perform hierarchical clustering processing on the target point cloud data, and using a LOF algorithm filtering processing to obtain target layered point cloud data of different layers of reinforcement mesh, and according to the target layered point cloud data, using the corresponding segmentation analysis algorithm to obtain pairs of anchor points for ranging, and fitting the pairs of anchor points to output target ranging data. The application improves the overall efficiency and segmentation accuracy of tunnel reinforcement mesh ranging, so as to improve the accuracy of tunnel reinforcement mesh data measurement.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of ranging based on point cloud data, and particularly relates to a tunnel reinforcement mesh ranging method and system, an electronic device, and a storage medium. BACKGROUND

[0002] The emergence of laser radar provides a convenient method for people to obtain spatial geometric structure data in the form of point cloud. Compared with ordinary two-dimensional image data, three-dimensional point cloud data can accurately and densely depict the overall structure of an object (such as the tunnel original scanning sample shown in FIG. 1), provide more rich spatial geometric structure information, and help understand the three-dimensional spatial characteristics of the environment and make corresponding decisions.

[0003] The measurement of tunnel reinforcement mesh related data (such as Figure 2 the layer-like spacing of the reinforcement mesh shown in FIG. 2, Figure 3 the lateral spacing between the reinforcement bars in the inner layer reinforcement mesh shown in FIG. 3, and Figure 4 the longitudinal spacing shown in FIG. 4) is an indispensable step for tunnel safety monitoring. Currently, laser radar scanning is usually used to obtain related point cloud data for tunnel reinforcement mesh. The obtained reinforcement mesh point cloud data is huge, consisting of millions or even billions of points in a large scene. However, the reinforcement mesh point cloud data has the characteristics of dispersion, irregularity, disorder, and uneven distribution. Using the segmentation and recognition calculation method of the prior art to process the reinforcement mesh point cloud data to calculate the measurement data such as the layer-like spacing of the reinforcement mesh, the lateral spacing between the reinforcement bars in the inner layer reinforcement mesh, and the longitudinal spacing often has defects such as insufficient real-time performance, difficulty in strictly mathematically describing the reinforcement mesh, and low signal-to-noise ratio, resulting in low overall efficiency and segmentation accuracy of tunnel reinforcement mesh ranging, thereby affecting the accuracy of tunnel reinforcement mesh data measurement. SUMMARY

[0004] To solve the above technical problems, the present application provides a tunnel reinforcement mesh ranging method, system, electronic device, and storage medium, which splits the complex ranging scene into a combination of single ranging targets by filtering and segmenting point cloud data, focuses only on the local characteristics of the measurement object for each ranging target, and discards other complex semantic information, thereby improving the overall efficiency and segmentation accuracy of tunnel reinforcement mesh ranging and improving the accuracy of tunnel reinforcement mesh data measurement.

[0005] In a first aspect, the present application provides a tunnel reinforcement mesh ranging method, comprising:

[0006] acquiring initial point cloud data of a tunnel by laser radar scanning;

[0007] preprocessing the initial point cloud data to obtain down-sampled point cloud data of the tunnel;

[0008] The target point cloud data of the steel mesh to be measured is obtained by filtering non-target point cloud data from the downsampled point cloud data using a preset algorithm, wherein the preset algorithm includes a ground feature recognition algorithm and an octree algorithm.

[0009] Combining the cross-section design data, an extended coordinate reflection algorithm is used to calculate the operational feature information between each point in the target point cloud data and its corresponding design point in the cross-section, wherein the operational feature information includes mileage information and deviation information;

[0010] Based on the business characteristic information, the target point cloud data is subjected to hierarchical clustering processing using the KMean clustering analysis method and filtered using the LOF algorithm to obtain target layered point cloud data of different layers of steel mesh.

[0011] Based on the target layered point cloud data, the corresponding segmentation analysis algorithm is used to obtain the pair of anchor points for distance measurement, and the pair of anchor points are fitted to output the target distance measurement data of the steel bar to be measured. The target distance measurement data includes the steel mesh layer spacing and the transverse and longitudinal spacing of a single layer of steel mesh.

[0012] Preferably, the step of preprocessing the initial point cloud data to obtain the downsampled point cloud data of the tunnel specifically includes:

[0013] Select the appropriate target downsampling method based on the downsampling method;

[0014] Based on the target downsampling method, the initial point cloud data is downsampled to transform the initial point cloud data into the downsampled point cloud data obtained by the downsampling process.

[0015] Preferably, the target downsampling method is a uniform downsampling method or a voxel downsampling method.

[0016] Preferably, the step of filtering non-target point cloud data from the downsampled point cloud data to obtain the target point cloud data of the steel mesh to be measured using a preset algorithm, wherein the preset algorithm includes a feature recognition algorithm and an octree algorithm, specifically comprising:

[0017] A ground feature recognition algorithm is used to extract the point cloud data of the tunnel surface from the downsampled point cloud data to obtain the retained point cloud data;

[0018] The point cloud neighborhood range of the retained point cloud data is calculated using the octree algorithm, and the retained point cloud data is filtered based on the point spacing within the point cloud neighborhood range to obtain the target point cloud data of the steel mesh to be measured.

[0019] Preferably, the step of combining cross-section design data and using an extended coordinate reflection algorithm to calculate the operational feature information between each point in the target point cloud data and its corresponding design point in its cross-section, wherein the operational feature information includes mileage information and deviation information, specifically includes:

[0020] Based on the preset line data, the line element closest to each point in the target point cloud data on the plane projection is calculated using XY coordinates;

[0021] The projection distance from the projection point of the corresponding line element to the endpoint of the line element is calculated based on the linear space.

[0022] The mileage information corresponding to each point in the target point cloud data is calculated based on the mileage of the endpoint of the line element and the projection distance;

[0023] Calculate the horizontal deviation of each point in the target point cloud data from the corresponding line element based on the coordinates of the projection points;

[0024] Based on the mileage information and the vertical curve information in the alignment data, the vertical offset of each point in the target point cloud data relative to the corresponding alignment is calculated.

[0025] Based on the cross-sectional design model, the two-dimensional position of the axis in the cross-section is determined, and complete business characteristic information is obtained.

[0026] Preferably, the step of performing hierarchical clustering of the target point cloud data using the KMean clustering analysis method and filtering using the LOF algorithm based on the business feature information to obtain target layered point cloud data of different layers of steel mesh specifically includes:

[0027] The target point cloud data is sliced ​​based on the mileage information to obtain point cloud slice data.

[0028] By performing KMean clustering analysis on the deviation information, the point cloud slice data is processed into layers to obtain layered point cloud data of different layers of steel mesh.

[0029] The LOF algorithm is used to filter out outliers in the layered point cloud data to obtain the target layered point cloud data.

[0030] Preferably, the segmentation analysis algorithm includes segmentation analysis based on the smoothness of deviation changes caused by the smoothness of the tunnel inner wall and the uniformity of the deviation gradient, which is beneficial to the layer spacing of the steel mesh, and segmentation analysis based on the obvious mesh characteristics of a single layer of steel mesh, which is beneficial to the transverse and longitudinal spacing of the single layer of steel mesh.

[0031] Secondly, the present invention provides a tunnel reinforcement mesh ranging system, comprising:

[0032] The acquisition module is used to acquire initial point cloud data of the tunnel by scanning with lidar;

[0033] The preprocessing module is used to preprocess the initial point cloud data to obtain the downsampled point cloud data of the tunnel;

[0034] The filtering module is used to filter out non-target point cloud data from the downsampled point cloud data to obtain the target point cloud data of the steel mesh to be measured by a preset algorithm, wherein the preset algorithm includes a ground feature recognition algorithm and an octree algorithm.

[0035] The calculation module is used to combine the cross-section design data and use an extended coordinate reflection algorithm to calculate the business feature information between each point in the target point cloud data and its corresponding design point in the cross section, wherein the business feature information includes mileage information and deviation information.

[0036] The hierarchical clustering module is used to perform hierarchical clustering processing on the target point cloud data using the KMean clustering analysis method and filtering processing using the LOF algorithm to obtain target layered point cloud data of different layers of steel mesh;

[0037] The segmentation analysis module is used to obtain paired anchor points for distance measurement based on the target layered point cloud data using the corresponding segmentation analysis algorithm, and to fit the paired anchor points to output the target distance measurement data of the steel bar to be measured, wherein the target distance measurement data includes the steel mesh layer spacing and the transverse and longitudinal spacing of a single layer of steel mesh.

[0038] Preferably, the preprocessing module includes:

[0039] The selection unit is used to select the appropriate target downsampling method based on the downsampling method.

[0040] The preprocessing unit is used to perform downsampling processing on the initial point cloud data based on the target downsampling method, and to transform the initial point cloud data into downsampled point cloud data obtained by the downsampling processing.

[0041] Preferably, the filtering module includes:

[0042] The stripping unit is used to strip the point cloud data of the tunnel surface from the downsampled point cloud data using a ground feature recognition algorithm to obtain the retained point cloud data;

[0043] The filtering module is used to calculate the neighborhood range of the retained point cloud data using an octree algorithm, and filter the retained point cloud data based on the point spacing within the neighborhood range to obtain the target point cloud data of the steel mesh to be measured.

[0044] Preferably, the computing module includes:

[0045] The first calculation unit is used to calculate the closest line element on the plane projection of each point in the target point cloud data based on the preset line data and the XY coordinates.

[0046] The second calculation unit is used to calculate the projection distance from the projection point of the point on the corresponding line element to the endpoint of the line element based on the linear space.

[0047] The third calculation unit is used to calculate the mileage information corresponding to each point in the target point cloud data based on the mileage of the endpoint of the line element and the projection distance;

[0048] The fourth calculation unit is used to calculate the horizontal deviation of each point in the target point cloud data from the corresponding line element based on the coordinates of the projection point;

[0049] The fifth calculation unit is used to calculate the vertical offset of each point in the target point cloud data relative to the corresponding alignment based on the mileage information and the vertical curve information in the alignment data.

[0050] The determination unit is used to determine the two-dimensional position of the axis in the cross section based on the cross section design model, thereby obtaining complete business characteristic information.

[0051] Preferably, the hierarchical clustering module includes:

[0052] A slicing unit is used to slice the target point cloud data according to the mileage information to obtain point cloud slice data.

[0053] The layered unit is used to perform layered processing on the point cloud slice data to obtain layered point cloud data of different layers of steel mesh by KMean clustering analysis for the deviation information.

[0054] The filtering unit is used to filter out outliers in the layered point cloud data using the LOF algorithm to obtain the target layered point cloud data.

[0055] Thirdly, the present invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the tunnel reinforcement mesh ranging method as described in the first aspect.

[0056] Fourthly, this embodiment provides a storage medium on which a computer program is stored, which, when executed by a processor, implements the tunnel reinforcement mesh ranging method as described in the first aspect.

[0057] Compared to existing technologies, this invention provides a method, system, electronic device, and storage medium for measuring the distance of a tunnel reinforcement mesh. The method involves: pre-processing the collected initial tunnel point cloud data by downsampling to obtain uniformly distributed downsampled point cloud data; filtering non-target point cloud data using a ground feature recognition algorithm and an octree algorithm to obtain target point cloud data for the reinforcement mesh to be measured; combining cross-section design data and employing an extended coordinate reflection algorithm to calculate the operational feature information between each point in the target point cloud data and its corresponding design point in the cross-section; using KMean clustering analysis to perform hierarchical clustering of the target point cloud data and using LOF algorithm filtering to obtain target layered point cloud data for different layers of the reinforcement mesh; and using corresponding segmentation analysis algorithms to obtain paired anchor points for distance measurement based on the target layered point cloud data, and fitting these paired anchor points to output the target distance measurement data for the reinforcement mesh to be measured. By filtering and segmenting point cloud data through the above steps, the complex ranging scenario is broken down into a combination of single ranging targets. For each ranging target, only the local characteristics of the measured object are considered, while other complex semantic information is discarded. This improves the overall efficiency and segmentation accuracy of tunnel rebar mesh ranging and enhances the accuracy of tunnel rebar mesh data measurement. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 The original scan sample after the 3D scanning of the tunnel;

[0060] Figure 2 This is a schematic diagram illustrating the layer spacing analysis of the tunnel reinforcement mesh.

[0061] Figure 3 A schematic diagram illustrating the analysis of the transverse spacing between the inner layers of steel bars in a tunnel steel mesh.

[0062] Figure 4 A schematic diagram illustrating the longitudinal spacing analysis between the inner layers of steel bars in a tunnel steel mesh.

[0063] Figure 5 This is a flowchart of the tunnel reinforcement mesh distance measurement method provided in Embodiment 1 of the present invention;

[0064] Figure 6 This is a statistical chart showing the overall measurement of the steel reinforcement layer spacing in the tunnel steel reinforcement mesh distance measurement method provided in Embodiment 1 of the present invention;

[0065] Figure 7 This is a statistical chart of the transverse rebar distance measurement of the tunnel rebar mesh distance measurement method provided in Embodiment 1 of the present invention;

[0066] Figure 8 This is a statistical chart of the longitudinal reinforcement distance measurement of the tunnel reinforcement mesh distance measurement method provided in Embodiment 1 of the present invention;

[0067] Figure 9 This is a structural block diagram of the tunnel steel mesh ranging system corresponding to the method in Embodiment 1, provided in Embodiment 2 of the present invention;

[0068] Figure 10 This is a schematic diagram of the hardware structure of the electronic device provided in Embodiment 3 of the present invention.

[0069] Explanation of reference numerals in the attached figures:

[0070] 10 - Data Acquisition Module;

[0071] 20 - Preprocessing module, 21 - Selected unit, 22 - Preprocessing unit;

[0072] 30 - Filtering module; 31 - Stripping unit; 32 - Filtering module;

[0073] 40 - Calculation module, 41 - First calculation unit, 42 - Second calculation unit, 43 - Third calculation unit, 44 - Fourth calculation unit, 45 - Fifth calculation unit, 46 - Determination unit;

[0074] 50 - Hierarchical clustering module, 51 - Slicing unit, 52 - Hierarchical unit, 53 - Screening unit;

[0075] 60-Segmentation Analysis Module;

[0076] 70-Bus, 71-Processor, 72-Memory, 73-Communication interface. Detailed Implementation

[0077] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0078] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0079] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0080] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0081] Example 1

[0082] Specifically, Figure 5 The diagram shown is a flowchart of a tunnel reinforcement mesh distance measurement method provided in this embodiment.

[0083] like Figure 5 As shown, the tunnel reinforcement mesh distance measurement method in this embodiment includes the following steps:

[0084] S101 uses lidar to scan and collect initial point cloud data of the tunnel.

[0085] Specifically, lidar is classified according to its scanning method into MEMS lidar, Flash lidar, phased array lidar, terrestrial lidar, and mechanically rotating lidar. In this embodiment, a terrestrial lidar is selected. The characteristics of a terrestrial lidar include: one-click vegetation filtering; analysis of multiple echo data with at least four categories; control of scanner operation, data processing, data export (LAS format point cloud data and image data), generation of contour lines, and cut / fill volume analysis; acquisition of 3D point cloud data with different point densities; acquisition of 3D color point cloud data; browsing of massive point cloud data; one-click generation of triangular mesh models with editable models; generation of cross-sectional lines; and classification and display of data echoes, with at least four categories. In practical application, this embodiment can use a terrestrial lidar to scan a highway tunnel to be measured, obtaining data such as... Figure 1 The original scan sample of the tunnel shown.

[0086] S102, preprocess the initial point cloud data to obtain the downsampled point cloud data of the tunnel.

[0087] Specifically, when using ground-based lidar to collect data, due to the high scanning accuracy and large single scanning range, the collected point cloud data exhibits characteristics such as non-uniformity, massive volume, and disorder. Direct processing not only requires high computational costs but also leads to analysis errors due to the inherent non-uniformity of the data.

[0088] Furthermore, S102 of this embodiment specifically includes:

[0089] S1021, Select the appropriate target downsampling method according to the downsampling method;

[0090] Specifically, the target downsampling methods include uniform downsampling and voxel downsampling. Uniform downsampling has several different sampling methods, among which farthest-point sampling is relatively simple. First, a seed point needs to be selected, and an interior point set needs to be set. Each time, a point is found from the set of non-interior points in the point cloud that is farthest from the interior point. The characteristics of farthest-point sampling include: uniform distribution of sampling points, relatively high algorithm time complexity, and sampling points are generally initially distributed near the boundary. Voxel downsampling refers to voxelizing the 3D space and then sampling a point within each voxel. The center point or the point closest to the center is usually used as the sampling point. The specific method is as follows: 1. Create voxels: Calculate the bounding box of the point cloud and then discretize the bounding box into smaller voxels. The length, width, and height of the voxels can be set by the user or obtained by setting the number of grid points in the three directions of the bounding box; 2. Each smaller voxel contains several points, and the center point or the point closest to the center point is taken as the sampling point. Voxel sampling is characterized by its high efficiency and relatively uniform distribution of sampling points, but its uniformity is not as high as that of uniform sampling. While the spacing between sampling points can be controlled by the size of the voxels, the number of sampling points cannot be precisely controlled. In this embodiment, to balance efficiency and data volume, a combination of uniform downsampling and voxel downsampling is used. Uniform downsampling is used at the data acquisition client to control the data volume before storing the point cloud in the database. In subsequent analysis and preprocessing, voxel downsampling is used for secondary processing to control the uniformity of the point cloud as much as possible.

[0091] S1022, Based on the target downsampling method, downsampling processing is performed on the initial point cloud data to convert the initial point cloud data into downsampled point cloud data obtained by the downsampling processing.

[0092] Specifically, in this embodiment, the common preprocessing measure before analysis is to downsample the point cloud, converting the operation on the entire point cloud to the points obtained by downsampling, so that the overall distribution of the sample is as uniform as possible and the total amount of computation is reduced.

[0093] S103, the target point cloud data of the steel mesh to be measured is obtained by filtering non-target point cloud data from the downsampled point cloud data through a preset algorithm, wherein the preset algorithm includes a ground feature recognition algorithm and an octree algorithm.

[0094] Specifically, the amount of steel mesh point cloud data collected from the real world is enormous, consisting of large scenes with millions or even billions of points, requiring high computational costs. Furthermore, unlike the simpler structure of traditional two-dimensional images, three-dimensional point cloud data containing steel mesh contains a large number of noisy points in three-dimensional space. The point cloud data itself exhibits: dispersion – the points in the point cloud are not connected to each other and exist scattered; irregularity – the scene requiring point cloud identification lacks a uniform shape and contains a large amount of noise; and uneven distribution – the point cloud density is unevenly distributed in space, with a trend of exponential decrease in point cloud density from the scanning point to distant locations.

[0095] Furthermore, S103 of this embodiment specifically includes:

[0096] S1031, a ground feature recognition algorithm is used to extract the point cloud data of the tunnel surface from the downsampled point cloud data to obtain the retained point cloud data.

[0097] Specifically, land cover recognition algorithms employ hyperspectral imaging technology. Spectral curve analysis typically involves difference analysis, using models such as minimum distance, spectral angle matching, and spectral similarity. In practical applications, the spectral differences between different types of land cover of the same type may be too small, causing curve similarity measurement methods to fail. In such cases, spectral curve feature analysis can be performed by focusing on characteristic points such as peaks, troughs, and inflection points. These feature points contain a wealth of information, determining the shape of the curve. By utilizing hyperspectral imaging for land cover recognition, point cloud data of the tunnel surface can be extracted from the downsampled point cloud data.

[0098] S1032, calculate the neighborhood range of the retained point cloud data using the octree algorithm, and filter the retained point cloud data based on the point spacing within the neighborhood range to obtain the target point cloud data of the steel mesh to be measured.

[0099] Specifically, the octree algorithm is an important tool for processing 3D point cloud data. Neighborhood search in an octree can achieve the function of finding neighbors with just a few lines of code. An octree has either 8 or 0 child nodes; the existence of empty nodes is because they do not contain any data points, thus improving traversal efficiency. In this embodiment, the octree algorithm can eliminate the influence of outliers in the point cloud data, further improving data accuracy.

[0100] S104, Combining the cross-section design data, an extended coordinate reflection algorithm is used to calculate the operational feature information between each point in the target point cloud data and its corresponding design point in its cross-section, wherein the operational feature information includes mileage information and deviation information.

[0101] Specifically, traditional design coordinate inverse calculation calculates the mileage, horizontal and vertical offset of each point in the point cloud by solving the horizontal and vertical curve information. This embodiment combines cross-section design data to simultaneously calculate the deviation of each point from its corresponding design point in its cross-section, as well as the point's position information within the cross-section design outline.

[0102] Furthermore, S104 of this embodiment specifically includes:

[0103] S1041, Based on the preset line data, calculate the closest line element on the plane projection of each point in the target point cloud data using XY coordinates;

[0104] S1042, calculate the projection distance from the projection point of the point on the corresponding line element to the endpoint of the line element based on the linear space.

[0105] S1043, calculate the mileage information corresponding to each point in the target point cloud data based on the mileage of the endpoint of the line element and the projection distance;

[0106] S1044, Calculate the horizontal deviation of each point in the target point cloud data from the corresponding line element based on the coordinates of the projection point;

[0107] S1045, based on the mileage information and combined with the vertical curve information in the alignment data, calculate the vertical offset of each point in the target point cloud data relative to the corresponding alignment.

[0108] S1046. Based on the cross-sectional design model, determine the two-dimensional position of the axis in the cross-section to obtain complete business characteristic information.

[0109] S105, based on the business characteristic information, the target point cloud data is subjected to hierarchical clustering processing using the KMean clustering analysis method and filtering processing using the LOF algorithm to obtain target layered point cloud data of different layers of steel mesh.

[0110] Specifically, since the steel mesh itself has a clear layered characteristic, the point cloud can be divided into several layers by directly performing KMean clustering analysis on the design deviation data, and only the outermost layers of data are retained (determined according to the number of steel mesh layers on site); among them, the outermost layer of steel mesh and the tunnel wall are closely attached and will be regarded as the same layer in the clustering.

[0111] Furthermore, S105 of this embodiment specifically includes:

[0112] S1051, The target point cloud data is sliced ​​according to the mileage information to obtain point cloud slice data.

[0113] Specifically, based on the design mileage information, the complete tunnel is sliced, and each slice is analyzed separately, which reduces the amount of calculation per operation and makes the characteristics of its business data easier to analyze in subsequent analyses.

[0114] S1052, by performing KMean clustering analysis on the deviation information, the point cloud slice data is processed into layers to obtain layered point cloud data of different layers of steel mesh.

[0115] Specifically, KMean clustering analysis divides the data into a specified k clusters, and the center point of each cluster is calculated from the mean of the samples in each cluster. The process involves continuously calculating the distance between each sample point and the cluster center until convergence.

[0116] S1053, The LOF algorithm is used to filter out outliers in the layered point cloud data to obtain the target layered point cloud data.

[0117] Specifically, the steel mesh data after initial stratification can be filtered a second time. Since the point cloud range shrinks again at this time, its statistical characteristics change significantly. The LOF algorithm can be used to further filter out outliers.

[0118] S106, according to the target layered point cloud data, the corresponding segmentation analysis algorithm is used to obtain the pair of anchor points for distance measurement, and the pair of anchor points are fitted to output the target distance measurement data of the steel bar to be measured, wherein the target distance measurement data includes the steel mesh layer spacing, the transverse and longitudinal spacing of the single layer steel mesh.

[0119] Specifically, different semantic recognition focuses exist when facing different scenarios:

[0120] 1. The layer spacing report focuses more on the distance between different layers, requiring accurate separation of the reinforcing mesh from the tunnel wall. The analytical indicators in the layer spacing report are: a) standard distance between reinforcing bars: 0.54-0.66 meters; b) a spacing exceeding 0.66 meters is considered excessive; c) a spacing less than 0.54 meters is considered insufficient. Specific statistical reports are as follows: Figure 6 As shown in the figure. In this embodiment, after the secondary lining construction, the tunnel inner wall is relatively smooth overall, and the deviation changes between each point on the tunnel inner wall and the design are also relatively smooth, with a relatively uniform gradient of deviation at each point. However, the steel mesh, due to its mesh structure, exhibits more abrupt changes. Utilizing this construction characteristic, the gradient of the design deviation at each point can be used as the basis for statistical filtering to perform inverse filtering on the point cloud, retaining abrupt change data, that is, retaining the steel mesh information as much as possible while removing the inner wall. The layer spacing measurement statistics table of this embodiment is shown in Table 1.

[0121] Table 1: Statistical Table of Inter-layer Spacing Measurement

[0122]

[0123] 2. Horizontal and vertical reports focus more on the semantic division within the same layer of steel mesh. Single-layer steel mesh has obvious mesh characteristics. Based on this characteristic, business data (mileage, horizontal deviation, vertical deviation) is used as a new local coordinate system. The point cloud is sliced ​​using octree, and the point cloud is sliced ​​using empirical values ​​of steel mesh gap size. Connectivity analysis is performed using octree, and the point cloud is grouped and fitted in the contour direction to obtain paired anchor points for distance measurement.

[0124] 2.1 Lateral Report Analysis Indicators: a) Standard distance between reinforcing bars: 0.15-0.18 meters; b) Reinforcing bar spacing exceeding 0.18 meters is considered excessive; c) Reinforcing bar spacing less than 0.15 meters is considered insufficient. The lateral reinforcing bar spacing statistics table for this embodiment is shown in Table 2; the specific statistical report is as follows... Figure 7 As shown.

[0125] Table 2: Statistics on the Measurement of Transverse Reinforcing Bars 2.2 Lateral Report Analysis Indicators: a) Standard distance between reinforcing bars: 0.20-0.26 meters; b) Reinforcing bar spacing exceeding 0.26 meters is considered excessive; c) Reinforcing bar spacing less than 0.20 meters is considered insufficient. The lateral reinforcing bar spacing statistics table in this embodiment is shown in Table 3; the specific statistical report is as follows... Figure 8 As shown.

[0126] Table 3: Statistics on Longitudinal Reinforcement Distance Measurement In summary, by filtering and segmenting the collected tunnel point cloud data through the above steps, the complex ranging scenario can be broken down into a combination of single ranging targets. For each ranging target, only the local characteristics of the measured object are considered, while other complex semantic information is discarded. This improves the overall efficiency and segmentation accuracy of tunnel rebar mesh ranging and enhances the accuracy of tunnel rebar mesh data measurement.

[0127] Example 2

[0128] This embodiment provides a structural block diagram of a system corresponding to the method described in Embodiment 1. Figure 9 This is a structural block diagram of the tunnel reinforcement mesh ranging system according to this embodiment, as follows: Figure 9 As shown, the system includes:

[0129] Acquisition module 10 is used to acquire initial point cloud data of the tunnel by scanning with lidar;

[0130] Preprocessing module 20 is used to preprocess the initial point cloud data to obtain downsampled point cloud data of the tunnel;

[0131] The filtering module 30 is used to filter out non-target point cloud data from the downsampled point cloud data to obtain the target point cloud data of the steel mesh to be measured by using a preset algorithm, wherein the preset algorithm includes a ground feature recognition algorithm and an octree algorithm.

[0132] The calculation module 40 is used to combine the cross-section design data and use an extended coordinate reflection algorithm to calculate the business feature information between each point in the target point cloud data and its corresponding design point in the cross section, wherein the business feature information includes mileage information and deviation information.

[0133] The hierarchical clustering module 50 is used to perform hierarchical clustering processing on the target point cloud data using the KMean clustering analysis method and to perform filtering processing using the LOF algorithm to obtain target layered point cloud data of different layers of steel mesh.

[0134] The segmentation analysis module 60 is used to obtain paired anchor points for distance measurement based on the target layered point cloud data using corresponding segmentation analysis algorithms, and to fit the paired anchor points to output the target distance measurement data of the reinforcing bar to be measured. The target distance measurement data includes the reinforcing mesh layer spacing, and the transverse and longitudinal spacing of a single-layer reinforcing mesh. In this embodiment, the segmentation analysis algorithm includes segmentation analysis based on the smoothness of deviation changes caused by the smoothness of the tunnel inner wall and the uniformity of the deviation gradient, which is beneficial to the reinforcing mesh layer spacing; and segmentation analysis based on the obvious mesh characteristics of a single-layer reinforcing mesh, which is beneficial to the transverse and longitudinal spacing of the single-layer reinforcing mesh.

[0135] Furthermore, the preprocessing module 20 includes:

[0136] Select unit 21 is used to select an appropriate target downsampling method based on the downsampling method;

[0137] Preprocessing unit 22 is used to perform downsampling processing on the initial point cloud data based on the target downsampling method, and to convert the initial point cloud data into downsampled point cloud data obtained by the downsampling processing.

[0138] Furthermore, the filtering module 30 includes:

[0139] The stripping unit 31 is used to strip the point cloud data of the tunnel surface from the downsampled point cloud data using a ground feature recognition algorithm to obtain the retained point cloud data;

[0140] The filtering module 32 is used to calculate the neighborhood range of the retained point cloud data using an octree algorithm, and filter the retained point cloud data based on the point spacing within the neighborhood range to obtain the target point cloud data of the steel mesh to be measured.

[0141] Furthermore, the computing module 40 includes:

[0142] The first calculation unit 41 is used to calculate the closest line element on the plane projection of each point in the target point cloud data according to the preset line data and the XY coordinates.

[0143] The second calculation unit 42 is used to calculate the projection distance from the projection point of the point on the corresponding line element to the endpoint of the line element based on the linear space.

[0144] The third calculation unit 43 is used to calculate the mileage information corresponding to each point in the target point cloud data based on the mileage of the endpoint of the line element and the projection distance;

[0145] The fourth calculation unit 44 is used to calculate the horizontal deviation of each point in the target point cloud data from the corresponding line element based on the coordinates of the projection point;

[0146] The fifth calculation unit 45 is used to calculate the vertical offset of each point in the target point cloud data relative to the corresponding alignment based on the mileage information and the vertical curve information in the alignment data.

[0147] Unit 46 is used to determine the two-dimensional position of the axis in the cross section based on the cross section design model, so as to obtain complete business feature information.

[0148] Furthermore, the hierarchical clustering module 50 includes:

[0149] Slicing unit 51 is used to slice the target point cloud data according to the mileage information to obtain point cloud slice data.

[0150] Layering unit 52 is used to perform layering processing on the point cloud slice data to obtain layered point cloud data of different layers of steel mesh by KMean clustering analysis for the deviation information;

[0151] The filtering unit 53 is used to filter out outliers in the layered point cloud data using the LOF algorithm to obtain the target layered point cloud data.

[0152] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0153] Example 3

[0154] Combination Figure 5 The described method for measuring distances using tunnel steel mesh can be implemented using electronic equipment. Figure 10This is a schematic diagram of the hardware structure of an electronic device according to this embodiment.

[0155] The electronic device may include a processor 71 and a memory 72 storing computer program instructions.

[0156] Specifically, the processor 71 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement this embodiment.

[0157] The memory 72 may include a mass storage device for data or instructions. For example, and not limitingly, the memory 72 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), flash memory, an optical disk drive, a magneto-optical disk drive, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 72 may include removable or non-removable (or fixed) media. Where appropriate, the memory 72 may be internal or external to a data processing device. In a particular embodiment, the memory 72 is non-volatile memory. In a particular embodiment, the memory 72 includes read-only memory (ROM) and random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), an electrically alterable read-only memory (EAROM), or flash memory, or a combination of two or more of these. Where appropriate, the RAM can be Static Random-Access Memory (SRAM) or Dynamic Random-Access Memory (DRAM). DRAM can be Fast Page Mode Dynamic Random-Access Memory (FPMDRAM), Extended Data Out Dynamic Random-Access Memory (EDODRAM), Synchronous Dynamic Random-Access Memory (SDRAM), etc.

[0158] The memory 72 can be used to store or cache various data files that need to be processed and / or communicated, as well as possible computer program instructions executed by the processor 71.

[0159] The processor 71 reads and executes the computer program instructions stored in the memory 72 to implement the tunnel steel mesh distance measurement method of Embodiment 1 described above.

[0160] In some embodiments, the electronic device may further include a communication interface 73 and a bus 70. For example, Figure 10 As shown, the processor 71, memory 72, and communication interface 73 are connected through bus 70 and complete communication with each other.

[0161] The communication interface 73 is used to enable communication between the various modules, devices, units, and / or equipment in this embodiment. The communication interface 73 can also enable data communication with other components such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations.

[0162] Bus 70 includes hardware, software, or both, that couples components of a device together. Bus 70 includes, but is not limited to, at least one of the following: data bus, address bus, control bus, expansion bus, and local bus. For example, and not as a limitation, bus 70 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 70 may include one or more buses. Although this embodiment describes and illustrates a specific bus, this application considers any suitable bus or interconnect.

[0163] The device can execute the tunnel rebar mesh ranging method of this embodiment 1 based on the tunnel rebar mesh ranging system.

[0164] Furthermore, in conjunction with the tunnel reinforcement mesh distance measurement method of Embodiment 1 above, this embodiment can provide a storage medium for implementation. This storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement the tunnel reinforcement mesh distance measurement method of Embodiment 1 above.

[0165] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0166] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for measuring distance using a tunnel steel mesh, characterized in that, include: Initial point cloud data of the tunnel was collected by LiDAR scanning; The initial point cloud data is preprocessed to obtain the downsampled point cloud data of the tunnel; The target point cloud data of the steel mesh to be measured is obtained by filtering non-target point cloud data from the downsampled point cloud data using a preset algorithm, wherein the preset algorithm includes a ground feature recognition algorithm and an octree algorithm. Combining the cross-section design data, an extended coordinate reflection algorithm is used to calculate the operational feature information between each point in the target point cloud data and its corresponding design point in the cross-section, wherein the operational feature information includes mileage information and deviation information; Based on the business characteristic information, the target point cloud data is subjected to hierarchical clustering processing using the KMean clustering analysis method and filtered using the LOF algorithm to obtain target layered point cloud data of different layers of steel mesh. Based on the target layered point cloud data, the corresponding segmentation analysis algorithm is used to obtain the pair of anchor points for distance measurement, and the pair of anchor points are fitted to output the target distance measurement data of the steel bar to be measured. The target distance measurement data includes the steel mesh layer spacing and the transverse and longitudinal spacing of a single layer of steel mesh.

2. The tunnel reinforcement mesh distance measurement method according to claim 1, characterized in that, The step of preprocessing the initial point cloud data to obtain the downsampled point cloud data of the tunnel specifically includes: Select the appropriate target downsampling method based on the downsampling method; Based on the target downsampling method, the initial point cloud data is downsampled to transform the initial point cloud data into the downsampled point cloud data obtained by the downsampling process.

3. The tunnel reinforcement mesh distance measurement method according to claim 2, characterized in that, The target downsampling method is specifically a uniform downsampling method or a voxel downsampling method.

4. The tunnel reinforcement mesh distance measurement method according to claim 1, characterized in that, The step of obtaining the target point cloud data of the steel mesh to be measured by filtering non-target point cloud data from the downsampled point cloud data using a preset algorithm, wherein the preset algorithm includes a ground feature recognition algorithm and an octree algorithm, specifically includes: A ground feature recognition algorithm is used to extract the point cloud data of the tunnel surface from the downsampled point cloud data to obtain the retained point cloud data; The point cloud neighborhood range of the retained point cloud data is calculated using the octree algorithm, and the retained point cloud data is filtered based on the point spacing within the point cloud neighborhood range to obtain the target point cloud data of the steel mesh to be measured.

5. The tunnel reinforcement mesh distance measurement method according to claim 1, characterized in that, The step of combining cross-section design data and using an extended coordinate reflection algorithm to calculate the operational feature information between each point in the target point cloud data and its corresponding design point in the cross-section, wherein the operational feature information includes mileage information and deviation information, specifically includes: Based on the preset line data, the line element closest to each point in the target point cloud data on the plane projection is calculated using XY coordinates; The projection distance from the projection point of the corresponding line element to the endpoint of the line element is calculated based on the linear space. The mileage information corresponding to each point in the target point cloud data is calculated based on the mileage of the endpoint of the line element and the projection distance; Calculate the horizontal deviation of each point in the target point cloud data from the corresponding line element based on the coordinates of the projection points; Based on the mileage information and the vertical curve information in the alignment data, the vertical offset of each point in the target point cloud data relative to the corresponding alignment is calculated. Based on the cross-sectional design model, the two-dimensional position of the axis in the cross-section is determined, and complete business characteristic information is obtained.

6. The tunnel reinforcement mesh distance measurement method according to claim 1, characterized in that, The steps of performing hierarchical clustering of the target point cloud data using the KMean clustering analysis method and filtering using the LOF algorithm based on the business feature information to obtain target layered point cloud data of different layers of steel mesh specifically include: The target point cloud data is sliced ​​based on the mileage information to obtain point cloud slice data. By performing KMean clustering analysis on the deviation information, the point cloud slice data is processed into layers to obtain layered point cloud data of different layers of steel mesh. The LOF algorithm is used to filter out outliers in the layered point cloud data to obtain the target layered point cloud data.

7. The tunnel reinforcement mesh distance measurement method according to claim 1, characterized in that, The segmentation analysis algorithm includes segmentation analysis based on the smoothness of deviation changes caused by the smoothness of the tunnel inner wall and the uniformity of the deviation gradient, which is beneficial to the layer spacing of the steel mesh, and segmentation analysis based on the obvious mesh characteristics of a single layer of steel mesh, which is beneficial to the transverse and longitudinal spacing of the single layer of steel mesh.

8. A tunnel steel mesh distance measuring system, characterized in that, include: The acquisition module is used to acquire initial point cloud data of the tunnel by scanning with lidar; The preprocessing module is used to preprocess the initial point cloud data to obtain the downsampled point cloud data of the tunnel; The filtering module is used to filter out non-target point cloud data from the downsampled point cloud data to obtain the target point cloud data of the steel mesh to be measured by a preset algorithm, wherein the preset algorithm includes a ground feature recognition algorithm and an octree algorithm. The calculation module is used to combine the cross-section design data and use an extended coordinate reflection algorithm to calculate the business feature information between each point in the target point cloud data and its corresponding design point in the cross section, wherein the business feature information includes mileage information and deviation information. The hierarchical clustering module is used to perform hierarchical clustering processing on the target point cloud data using the KMean clustering analysis method and filtering processing using the LOF algorithm to obtain target layered point cloud data of different layers of steel mesh; The segmentation analysis module is used to obtain paired anchor points for distance measurement based on the target layered point cloud data using the corresponding segmentation analysis algorithm, and to fit the paired anchor points to output the target distance measurement data of the steel bar to be measured, wherein the target distance measurement data includes the steel mesh layer spacing and the transverse and longitudinal spacing of a single layer of steel mesh.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the tunnel reinforcement mesh distance measurement method as described in any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the tunnel reinforcement mesh distance measurement method as described in any one of claims 1 to 7.

Images