Method for measuring exposed parameters of a roadway bolt and related equipment

Abstract

This application provides a method and related equipment for measuring the exposed parameters of roadway anchor bolts. The method includes acquiring three-dimensional point cloud data within the roadway, and determining point cloud sub-data corresponding to each anchor bolt based on the three-dimensional point cloud data; for each anchor bolt's corresponding point cloud sub-data, performing axis enhancement extraction on the point cloud sub-data to obtain a target axis reference point and a target axis unit direction vector corresponding to the anchor bolt; based on the point cloud sub-data, the target axis reference point, and the target axis unit direction vector, determining the exposed start point and exposed end point of the anchor bolt; and based on the target axis reference point, the target axis unit direction vector, the exposed start point, and the exposed end point, determining the exposed parameters corresponding to the anchor bolt. This method solves the technical problem of inaccurate measurement of the exposed parameters of roadway anchor bolts in the prior art, and improves the measurement accuracy of the exposed parameters of roadway anchor bolts.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and related equipment for measuring the exposed parameters of roadway anchor bolts. Background Technology

[0002] Roadways are various passages drilled between the surface and the ore body, used for ore transportation, ventilation, drainage, pedestrian access, and various necessary preparatory works for new ore extraction by metallurgical equipment. Anchor bolts are rod-shaped components used for reinforcing and supporting the surrounding rock of roadways. Anchor bolts consist of a deeply embedded anchoring section within the ore body and an exposed section. The deeply embedded section is firmly rooted within the roadway rock wall, while the exposed section remains within the roadway space. Its exposed length and installation deflection directly affect the roadway support effect, construction quality acceptance, and underground operational safety assessment. Therefore, the accuracy of measuring the exposed parameters of the anchor bolts is crucial.

[0003] Traditional anchor bolt inspection methods mainly rely on manual visual inspection, measurement and recording, or local contact measurement. These methods suffer from problems such as high labor intensity, low inspection efficiency, strong subjectivity, and poor consistency of results. They are difficult to meet the needs of long-distance tunnels, large-scale, and continuous inspection, which in turn affects the measurement accuracy of the exposed length and radial deflection angle of the anchor bolt. Summary of the Invention

[0004] In view of this, the purpose of this application is to propose a method and related equipment for measuring the exposed parameters of roadway anchor bolts, so as to overcome all or part of the deficiencies in the prior art.

[0005] To achieve the above objectives, this application provides a method for measuring the exposed parameters of roadway anchor bolts, comprising: acquiring three-dimensional point cloud data within the roadway, and determining point cloud sub-data corresponding to each anchor bolt based on the three-dimensional point cloud data; performing axis enhancement extraction on the point cloud sub-data corresponding to each anchor bolt to obtain a target axis reference point and a target axis unit direction vector corresponding to the anchor bolt; determining the exposed start point and exposed end point corresponding to the anchor bolt based on the point cloud sub-data, the target axis reference point, and the target axis unit direction vector; and determining the exposed parameters corresponding to the anchor bolt based on the target axis reference point, the target axis unit direction vector, the exposed start point, and the exposed end point.

[0006] Optionally, determining the point cloud sub-data corresponding to each anchor rod based on the three-dimensional point cloud data includes: preprocessing the three-dimensional point cloud data, classifying the preprocessed three-dimensional point cloud data using a pre-trained point cloud semantic segmentation model to obtain anchor rod-type point cloud data; and clustering the anchor rod-type point cloud data to obtain point cloud sub-data corresponding to each anchor rod.

[0007] Optionally, the step of performing axis enhancement extraction on the point cloud sub-data to obtain the target axis reference point and the target axis unit direction vector corresponding to the anchor bolt includes: performing axis fitting on the point cloud sub-data to obtain an initial axis reference point and an initial axis unit direction vector; performing axial analysis and extraction on the point cloud sub-data along the initial axis unit direction vector to obtain the main trunk center point cloud sub-data corresponding to the anchor bolt; performing data enhancement on the main trunk center point cloud sub-data; and performing axis fitting on the data-enhanced main trunk center point cloud sub-data to obtain the target axis reference point and the target axis unit direction vector.

[0008] Optionally, determining the exposed start and end points of the anchor bolt based on the point cloud sub-data, the target axis reference point, and the target axis unit direction vector includes: projecting the point cloud sub-data along the target axis unit direction vector to obtain a one-dimensional projection sequence containing multiple projection values; discretizing the one-dimensional projection sequence to obtain an interval sequence containing multiple axial intervals; determining the corresponding weighted point density of each axial interval in the interval sequence based on the target axis reference point and the target axis unit direction vector; performing an exposure judgment on each axial interval in the interval sequence based on the corresponding weighted point density of each axial interval to obtain at least one candidate exposure interval; and for each candidate... The exposed section is defined by: determining the local axis reference point corresponding to the candidate exposed section; determining the average direction deviation angle of the candidate exposed section relative to the target axis unit direction vector based on the target axis unit direction vector and the local axis reference point; selecting at least one exposed section from all candidate exposed sections based on the average direction deviation angle corresponding to each candidate exposed section; determining the exposed starting point of the anchor rod from the wall based on the target axis reference point, the target axis unit direction vector, the weighted point density corresponding to each axial section, and the average direction deviation angle corresponding to each candidate exposed section; and finding the maximum projection value among all exposed sections, and determining the maximum projection value as the exposed ending point corresponding to the anchor rod.

[0009] Optionally, the step of determining the exposure of each axial interval in the interval sequence based on the corresponding weighted point density of each axial interval to obtain at least one candidate exposed interval includes: determining the change in weighted point density of each pair of adjacent axial intervals in the interval sequence based on the weighted point density of each axial interval; determining a density change threshold based on the weighted point density of all axial intervals; and for each pair of adjacent axial intervals, in response to determining that the change in weighted point density is less than or equal to the density change threshold, designating each axial interval corresponding to the pair of adjacent axial intervals as a candidate exposed interval.

[0010] Optionally, the step of selecting at least one exposed interval from all candidate exposed intervals based on the average directional deviation angle corresponding to each candidate exposed interval includes: for each candidate exposed interval, in response to determining that the average directional deviation angle corresponding to the candidate exposed interval is less than or equal to the directional consistency threshold, determining the candidate exposed interval as the exposed interval.

[0011] Optionally, determining the exposed starting point of the anchor bolt against the wall based on the target axis reference point, the target axis unit direction vector, the corresponding weighted point density of each axial interval, and the average direction deviation angle corresponding to each candidate exposed interval includes: for each axial interval, determining the weighted average off-axis scale corresponding to the axial interval based on the target axis reference point and the target axis unit direction vector; finding the weighted average off-axis scale corresponding to each exposed interval among the weighted average off-axis scales corresponding to all axial intervals, and calculating the mean of the weighted average off-axis scales corresponding to all exposed intervals; determining the stable rod-shaped reference scale based on the mean; and in each In the weighted point density corresponding to each axial interval, the weighted point density corresponding to each exposed interval is found, and in the average directional deviation angle corresponding to each candidate exposed interval, the average directional deviation angle corresponding to each exposed interval is found; for each exposed interval, based on the weighted point density corresponding to the exposed interval, the average directional deviation angle corresponding to the exposed interval, and the stable rod-shaped reference scale, the stability score corresponding to the exposed interval is determined; in response to determining that the stability score is greater than or equal to the interval score threshold, the exposed interval is taken as the target axial interval; the minimum projection value is found in all target axial intervals, and the minimum projection value is determined as the exposed starting point.

[0012] Optionally, determining the stability score corresponding to the exposed interval based on the weighted point density corresponding to the exposed interval, the average directional deviation angle corresponding to the exposed interval, and the stable rod-shaped reference scale includes: normalizing the weighted point density corresponding to the exposed interval to obtain a normalized value of the weighted point density; determining the average directional deviation angle score based on the average directional deviation angle corresponding to the exposed interval; determining the off-axis scale consistency score based on the weighted average off-axis scale and the stable rod-shaped reference scale corresponding to the exposed interval; and determining the stability score based on the normalized value of the weighted point density, the average directional deviation angle score, and the off-axis scale consistency score.

[0013] Optionally, determining the exposed parameters of the anchor bolt based on the target axis reference point, the target axis unit direction vector, the exposed start point, and the exposed end point includes: acquiring the axis data of the roadway, determining the data with the smallest distance to the exposed start point as the nearest point; acquiring other data within a predetermined range from the axis data that are within the nearest point; determining the local roadway axis direction based on the nearest point and the other data; determining the roadway cross-section reference plane based on the local roadway axis direction; projecting the target axis unit direction vector onto the roadway cross-section reference plane to obtain an in-plane projection vector, and normalizing the in-plane projection vector to obtain a unit radial vector; determining the radial deflection angle corresponding to the anchor bolt based on the angle between the target axis unit direction vector and the unit radial vector; determining the exposed length of the anchor bolt based on the exposed start point and the exposed end point; and using the radial deflection angle and the exposed length corresponding to the anchor bolt as the exposed parameters corresponding to the anchor bolt.

[0014] Based on the same inventive concept, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the method described above when executing the computer program.

[0015] As can be seen from the above, the method and related equipment for measuring the exposed parameters of roadway anchor bolts provided in this application include acquiring three-dimensional point cloud data within the roadway and determining the point cloud sub-data corresponding to each anchor bolt based on the three-dimensional point cloud data. This not only reduces interference from irrelevant data but also accurately distinguishes individual anchor bolts, avoiding confusion among multiple anchor bolt data. For the point cloud sub-data corresponding to each anchor bolt, axis enhancement extraction is performed on the point cloud sub-data to obtain the target axis reference point and the target axis unit direction vector corresponding to the anchor bolt. By performing axis enhancement extraction on a subset of the point cloud of a single anchor bolt, it is possible to filter out messy noise points and distorted points, accurately locate the center direction of the anchor bolt, clarify the anchor bolt axis reference point and direction vector, and weaken the influence of rock wall attachments and point cloud incompleteness, accurately restoring the actual posture of the anchor bolt. Based on the point cloud sub-data, the target axis reference point, and the target axis unit direction vector, the exposed start point and exposed end point corresponding to the anchor bolt are determined. By relying on the target axis reference point and the target axis unit direction vector to constrain the overall orientation, and without depending on local scattered point cloud morphology, even with obstructions or missing points in the local point cloud, it can still complete interval division and feature discrimination according to the inherent axial pattern of the anchor rod. It avoids misjudging boundary positions due to local data anomalies, thus stably locking the standard wall intersection point, effectively reducing judgment errors and improving the accuracy of exposed start and end point identification. Based on the target axis reference point, the target axis unit direction vector, the exposed start point, and the exposed end point, the corresponding exposed parameters of the anchor rod are determined, effectively reducing position determination and angle calculation errors and significantly improving the accuracy of exposed parameters. Attached Figure Description

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

[0017] Figure 1 This is a flowchart illustrating the method for measuring the exposed parameters of roadway anchor bolts according to an embodiment of this application.

[0018] Figure 2 This is an overall point cloud map of the three-dimensional point cloud data within the tunnel in this application embodiment; Figure 3 This is a partial point cloud map of the three-dimensional point cloud data within the tunnel in an embodiment of this application; Figure 4 This is a point cloud diagram of the point cloud sub-data of a single anchor bolt in an embodiment of this application; Figure 5 This is a diagram showing the refined anchor bolt axis of an embodiment of this application; Figure 6 This is a multi-index curve diagram of the stability of the exposed interval in an embodiment of this application; Figure 7 A parameter curve is used to determine the exposed starting point in the embodiments of this application; Figure 8 This is a diagram showing the radial deflection angle measurement results of an embodiment of this application; Figure 9 This is a diagram showing the measurement results of the exposed start point, exposed end point, and exposed length in an embodiment of this application. Figure 10 This is a diagram showing the overall exposed length measurement and radial deflection angle results of an embodiment of this application; Figure 11 This is a flowchart illustrating a method for measuring exposed parameters of roadway anchor bolts according to another embodiment of this application. Figure 12 This is a schematic diagram of the structure of the measuring device for the exposed parameters of roadway anchor bolts according to an embodiment of this application; Figure 13 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0020] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0021] As described in the background section, roadways are various passages drilled between the surface and the ore body, used for ore transportation, ventilation, drainage, pedestrian access, and various necessary preparatory works for new ore extraction by metallurgical equipment. Anchor bolts are rod-shaped components used for the reinforcement and support of the surrounding rock in roadways. Anchor bolts consist of a deeply buried anchoring section embedded in the ore body and an exposed section. The deeply buried section is firmly rooted deep into the roadway rock wall, while the exposed section remains within the roadway space. Its exposed length and installation deflection directly affect the roadway support effect, construction quality acceptance, and underground operational safety assessment. Therefore, the accuracy of measuring the exposed parameters of the anchor bolts is crucial.

[0022] Traditional anchor bolt detection methods mainly rely on manual visual inspection, ruler measurement, or localized contact measurement, which suffers from high labor intensity, low detection efficiency, strong subjectivity, and poor result consistency, making it difficult to meet the needs of long-distance tunnels, large-scale, and continuous detection. With the development of 3D laser scanning and point cloud processing technology, using point clouds to achieve automatic anchor bolt identification and measurement of exposed parameters has become an important development direction for intelligent detection in underground engineering. However, in the measured point clouds of underground tunnels, anchor bolt point clouds are easily affected by factors such as rough surrounding rock surfaces, shotcrete layers, steel mesh, pipelines, attachments, and local obstructions, often resulting in root missing points, uneven point density, end dispersion, and local missegmentation. This makes it difficult to stably determine the exposed start and end points of the anchor bolt, thus affecting the measurement accuracy of the exposed length and radial deflection angle of the anchor bolt.

[0023] Existing technologies are prone to instability in local wall geometry when there are significant wall undulations, numerous local pits, incomplete wall point cloud extraction, or occlusion interference near the root. This instability affects the reliability of determining the exposed starting point and further reduces the stability of the exposed length measurement results. Furthermore, existing methods do not adequately consider the continuity, directional consistency, and rod-scale stability of the exposed anchor section. In complex scenarios, their ability to identify the effective exposed section of the anchor is limited, making it difficult to balance measurement accuracy with engineering applicability.

[0024] In summary, existing technologies still suffer from insufficient adaptability to complex scenarios, low stability in determining the exposed starting point, and limited accuracy in measuring exposed parameters. There is an urgent need to propose a method that is suitable for complex underground point cloud scenarios, can reduce the dependence on local wall fitting, stably determine the exposed starting point, and automatically measure the exposed length and radial deflection angle of anchor bolts.

[0025] In view of this, embodiments of this application propose a method for measuring the exposed parameters of roadway anchor bolts, referring to... Figure 1 This includes the following steps: Step 101: Obtain three-dimensional point cloud data in the tunnel, and determine the point cloud sub-data corresponding to each anchor rod based on the three-dimensional point cloud data.

[0026] In this step, this application proposes a method for measuring the exposed parameters of roadway anchor bolts based on the identification of exposed sections and the adaptive determination of exposed starting points. This method uses the measured three-dimensional point cloud data of underground roadways as the data basis. After obtaining the three-dimensional point cloud data of anchor bolts, it does not focus on restoring the layout relationship of the anchor bolt group, but instead focuses on the enhanced extraction of the main structure of a single anchor bolt, the identification of stable exposed sections, the adaptive determination of the exposed starting points, the measurement of the exposed length, and the calculation of the radial deflection angle. Thus, under complex conditions of occlusion, root missing, uneven point density, interference from attachments, and local missegmentation, it can achieve stable measurement of the exposed parameters of anchor bolts.

[0027] First, 3D point cloud data of the tunnel is acquired. This 3D point cloud data is obtained by actually scanning the underground tunnel using a 3D laser scanning device. For example, the 3D point cloud data of the tunnel includes the 3D point cloud data of the surrounding rock, anchor bolts, and other auxiliary components. Each point in the 3D point cloud data contains 3D spatial coordinate information, which can be denoted as: (1) in, For the three-dimensional point cloud data within the tunnel, The total number of points in the 3D point cloud data. For the first in the 3D point cloud data The spatial coordinates of the points This represents the three-dimensional spatial coordinates of the point with index i.

[0028] Because the 3D point cloud data within the roadway is a mixed point cloud of multiple entities within the roadway, such as surrounding rock, anchor bolts, pipelines, steel mesh, and shotcrete layers, etc., in order to accurately determine the exposed parameters of each anchor bolt, based on the 3D point cloud data, the corresponding point cloud sub-data for each anchor bolt is determined. This not only reduces interference from irrelevant data but also accurately distinguishes individual anchor bolts, avoiding confusion among multiple anchor bolt data. This application does not primarily aim to restore the layout relationship of the anchor bolt group, but rather focuses on the accurate extraction and geometric feature analysis of the point cloud of individual anchor bolts, avoiding the failure of individual anchor bolt identification due to the invalidation of the layout pattern of the anchor bolt group.

[0029] Step 102: For the point cloud sub-data corresponding to each anchor bolt, perform axis enhancement extraction on the point cloud sub-data to obtain the target axis reference point and the target axis unit direction vector corresponding to the anchor bolt.

[0030] In this step, for the point cloud sub-data corresponding to each anchor bolt, axis enhancement extraction is performed on the point cloud sub-data to obtain the target axis reference point and the target axis unit direction vector corresponding to the anchor bolt. The target axis reference point can be a benchmark positioning point on the anchor bolt axis, and the target axis unit direction vector can be used to characterize the spatial orientation and tilt attitude of the anchor bolt axis. By performing axis enhancement extraction on a subset of the point cloud for a single anchor bolt, cluttered noise and distorted points can be filtered out, accurately locating the center direction of the anchor bolt. This not only clarifies the anchor bolt axis reference point and direction vector but also weakens the influence of rock wall attachments and point cloud incompleteness, accurately restoring the actual attitude of the anchor bolt. This provides a reliable benchmark for subsequent calculation of exposed length and determination of deflection angle, effectively improving the accuracy and stability of exposed parameter detection.

[0031] Step 103: Based on the point cloud sub-data, the target axis reference point, and the target axis unit direction vector, determine the exposed start point and exposed end point of the anchor bolt.

[0032] In this step, the point cloud sub-data records the three-dimensional spatial coordinates of the exposed portion of the anchor bolt, fully preserving its geometric shape, diameter, and end information. The target axis reference point is used to calibrate the center position of the anchor bolt, providing a benchmark for determining the axial interval and solving for the exposed start and end points. The target axis unit direction vector is used to characterize the extension direction of the anchor bolt body, serving as a unified directional basis for dividing the point cloud projection intervals and selecting the exposed intervals. Therefore, based on the point cloud sub-data, the target axis reference point, and the target axis unit direction vector, the corresponding exposed start and end points of the anchor bolt are determined. The exposed start point is the point where the anchor bolt body penetrates the wall from the rock mass, and the exposed end point is the end position of the exposed section of the anchor bolt away from the surrounding rock or wall surface. The intersection points on both sides of the bolt body are determined on the tunnel rock wall, with one intersection point being the exposed start point and the other intersection point being the exposed end point.

[0033] By relying on the reference point of the target axis and the unit direction vector of the target axis to constrain the overall direction, without relying on the local scattered point cloud shape, even if there are debris obstructions or local point cloud missing, it can still complete the interval division and feature discrimination according to the inherent axial law of the anchor rod. It will not misjudge the boundary position due to local data anomalies, thus stably locking the standard wall intersection point, effectively reducing the judgment error and improving the accuracy of exposed start and end point identification.

[0034] Step 104: Based on the target axis reference point, the target axis unit direction vector, the exposed start point, and the exposed end point, determine the exposed parameters corresponding to the anchor bolt.

[0035] In this step, the target axis reference point provides the center positioning benchmark for the anchor bolt, the unit direction vector of the target axis reflects the actual tilt posture of the anchor bolt, and the exposed start and end points define the accurate exposed boundaries at the wall. These four data points—anchor bolt spatial position, tilt orientation, and exposed range—provide the geometric conditions required for calculating length and angle. Therefore, based on the target axis reference point, the target axis unit direction vector, and the exposed start and end points, the corresponding exposed parameters of the anchor bolt are determined. These parameters include the radial deflection angle and the exposed length. Using the anchor bolt's center position and actual extension direction as the core basis, combined with the precisely defined exposed start and end boundaries, all calculations are based on the anchor bolt's own true spatial shape, unaffected by surrounding irrelevant point cloud data. This effectively reduces errors in position determination and angle measurement, significantly improving the accuracy of the exposed parameters.

[0036] Compared to existing technologies, this application no longer relies excessively on the geometric fitting of uneven local walls, significantly mitigating the interference caused by incomplete wall point clouds and root occlusion, thus improving the stability of exposed starting point determination from the root. Simultaneously, by constraining the continuity, directional consistency, and dimensional characteristics of the exposed anchor bolt sections with a unified axial direction, it enhances the ability to identify effective exposed sections in complex underground scenarios. This solves the shortcomings of traditional methods, such as poor scenario adaptability and large parameter measurement fluctuations, while also providing stable and accurate measurement data output under various complex working conditions. It balances measurement and calculation accuracy with practical application value in on-site engineering, meeting the actual needs of automated detection and quality verification in underground tunnel support engineering.

[0037] The above scheme acquires three-dimensional point cloud data within the tunnel and, based on this data, determines the point cloud sub-data corresponding to each anchor bolt. This not only reduces interference from irrelevant data but also accurately distinguishes individual anchor bolts, avoiding confusion between multiple anchor bolt data. For each anchor bolt's corresponding point cloud sub-data, axis enhancement extraction is performed to obtain the target axis reference point and target axis unit direction vector. By performing axis enhancement extraction on a subset of the point cloud data for a single anchor bolt, cluttered noise and distorted points can be filtered out, accurately locking the anchor bolt's center direction. This not only clarifies the anchor bolt's axis reference point and direction vector but also mitigates the influence of rock wall attachments and point cloud incompleteness, accurately restoring the actual anchor bolt posture. Based on the point cloud sub-data, the target axis reference point, and the target axis unit direction vector, the exposed start and end points of the anchor bolt are determined. By relying on the target axis reference point and the target axis unit direction vector to constrain the overall orientation, and without depending on local scattered point cloud morphology, even with obstructions or missing points in the local point cloud, it can still complete interval division and feature discrimination according to the inherent axial pattern of the anchor rod. It avoids misjudging boundary positions due to local data anomalies, thus stably locking the standard wall intersection point, effectively reducing judgment errors and improving the accuracy of exposed start and end point identification. Based on the target axis reference point, the target axis unit direction vector, the exposed start point, and the exposed end point, the corresponding exposed parameters of the anchor rod are determined, effectively reducing position determination and angle calculation errors and significantly improving the accuracy of exposed parameters.

[0038] In some embodiments, determining the point cloud sub-data corresponding to each anchor rod based on the three-dimensional point cloud data includes: preprocessing the three-dimensional point cloud data; classifying the preprocessed three-dimensional point cloud data using a pre-trained point cloud semantic segmentation model to obtain anchor rod-type point cloud data; and clustering the anchor rod-type point cloud data to obtain point cloud sub-data corresponding to each anchor rod.

[0039] In this embodiment, the three-dimensional point cloud data within the tunnel is a mixed point cloud including the tunnel walls, surrounding rock, anchor bolts, and other auxiliary components. For example... Figure 2 As shown, the overall point cloud can reflect the overall distribution of various structures within the tunnel space; such as Figure 3 As shown, the local point cloud further illustrates the spatial relationships between the anchor bolt's adjacent wall surface, the root transition zone, and the surrounding background points. Figure 2 and Figure 3As can be seen, there are mutual proximity, local occlusion, and background interference between the anchor bolt point cloud, wall point cloud, and local attachment points. Therefore, it is necessary to preprocess the 3D point cloud data and extract anchor bolt-like point clouds to reduce the influence of irrelevant scene point clouds on subsequent single anchor bolt identification, axis fitting, and exposed parameter measurement. For example, voxel downsampling is performed on the 3D point cloud data to reduce the number of points while retaining the main geometric structure; then, a statistical outlier removal method is used to remove isolated noise points, resulting in preprocessed 3D point cloud data.

[0040] Preprocessed 3D point cloud data is input into a pre-trained point cloud semantic segmentation model to perform semantic classification on various types of point cloud data in the alleyway scene, obtaining point cloud data of each type. From each type of point cloud data, anchor bolt type point cloud data is extracted, denoted as . ,in, This is a collection of point cloud data for anchor bolts. For preprocessed 3D point cloud data, This refers to any point cloud data from the preprocessed 3D point cloud data. The anchor bolt-related point cloud is separated from the surrounding rock, equipment, pipelines, and other interfering components in complex tunnel scenes, providing a data foundation for subsequent single anchor bolt extraction and exposed parameter calculation.

[0041] It should be noted that before clustering the anchor bolt point cloud data, the anchor bolt point cloud data can be verified by fitting the curved surface of the tunnel wall and analyzing the morphology of the protrusions. This process removes wall points, wall protrusions and debris, and interference points attached to the wall, accurately purifying the valid points that truly belong to the anchor bolts and reducing misjudgments caused by the wall structure and wall protrusions.

[0042] Since multiple anchor bolts typically exist simultaneously in underground tunnels, it is necessary to analyze the point cloud data of anchor bolts. Further instance partitioning is performed, decomposing it into multiple independent point cloud sub-data sets corresponding to individual anchor bolts. The anchor bolt-type point cloud data is then clustered to obtain the point cloud sub-data sets corresponding to each anchor bolt. Euclidean distance is used for clustering; during the clustering process, if any two point cloud data sets... and The Euclidean distance between them satisfies: (2) Then the two points are determined to belong to the same cluster candidate set, where, and This represents the three-dimensional spatial coordinates of the point with index a. and This represents the three-dimensional spatial coordinates of the point with index b. This is the clustering distance threshold, used to control the spatial connectivity determination of point clouds belonging to the same anchor bolt. Its value can range from 0.01 to 0.05 m. (3). Let the number of identified anchor bolts be... Then it can be expressed as: ,in, For the first The point cloud sub-data corresponding to each anchor bolt. For example... Figure 4 As shown, Figure 4 This is a point cloud diagram of the point cloud sub-data of a single anchor bolt in an embodiment of this application. Figure 4 The target anchor cluster in the text refers to a cluster composed of point cloud sub-data of a single anchor. Figure 4 As can be seen, the target anchor cluster exhibits a slender point cloud distribution characteristic extending along the direction of the anchor body, and is separated from other candidate clusters or background point clouds. This provides a clear object basis for subsequent single anchor axis fitting, trunk structure extraction, and exposed parameter calculation.

[0043] It should be noted that before fitting the axis to the point cloud sub-data, the point cloud sub-data corresponding to each anchor rod is verified by combining indicators such as the number of points, axial span, slenderness, and spatial continuity of the point cloud cluster. By comprehensively judging multiple features such as the number of points contained in the point cluster, the overall axial extension length, the proportion of the overall slender shape, and the degree of spatial continuity, invalid interference clusters with too few points, short and thick shape, and spatial discontinuity are eliminated. The effective target anchor rod clusters that meet the slender entity characteristics of the anchor rod are accurately selected, and the exclusive point cloud data of each independent anchor rod is accurately separated.

[0044] Clustering the point cloud data of anchor bolts yields point cloud sub-data for each anchor bolt, allowing for the direct separation of independent data for a single bolt from the mixed anchor bolt point cloud. This avoids identification failures that may result from relying on the layout patterns of anchor bolt groups, and provides a clear and crosstalk-free dedicated data foundation for subsequent axis fitting and exposed parameter calculation of individual anchor bolts. At the same time, it effectively reduces the overall computational load of data processing and improves the stability and accuracy of subsequent parameter calculations.

[0045] In some embodiments, the step of performing axis enhancement extraction on the point cloud sub-data to obtain the target axis reference point and the target axis unit direction vector corresponding to the anchor bolt includes: performing axis fitting on the point cloud sub-data to obtain an initial axis reference point and an initial axis unit direction vector; performing axial analysis and extraction on the point cloud sub-data along the initial axis unit direction vector to obtain the trunk center point cloud sub-data corresponding to the anchor bolt; performing data enhancement on the trunk center point cloud sub-data; and performing axis fitting on the data-enhanced trunk center point cloud sub-data to obtain the target axis reference point and the target axis unit direction vector.

[0046] In this embodiment, the point cloud sub-data for each anchor rod Axis fitting is performed; specifically, based on the point cloud sub-data of the anchor bolt, the initial axis reference point corresponding to the anchor bolt is determined. The geometric center of the anchor bolt point cloud is calculated using the following formula: (4), where, For the first The geometric center of the point cloud corresponding to each anchor bolt. Representing point cloud sub-data The number of points in the middle, For belonging to the first A point in the point cloud sub-data of an anchor bolt. This represents the anchor bolt point cloud data set. The geometric center of the point cloud corresponding to the anchor bolt is used as the initial axis reference point.

[0047] Based on the point cloud sub-data of the anchor bolts and the initial axis reference point, the covariance matrix of the point cloud is determined using the following formula: (5) Eigenvalue decomposition is performed on the covariance matrix of the point cloud to obtain multiple eigenvalues ​​and eigenvectors for each eigenvalue. Let the eigenvalues ​​satisfy: (6), among which, Represents the first eigenvalue. Represents the second eigenvalue. Represents the third eigenvalue. Then the largest eigenvalue corresponding feature vector satisfy: (7) Since anchor point clouds typically exhibit the strongest linear distribution along the axial direction, the largest eigenvalue is selected. corresponding feature vector The principal direction vector of the anchor rod is denoted as... The eigenvector corresponding to the largest eigenvalue among multiple eigenvalues ​​is normalized to obtain the initial axis unit direction vector. To facilitate subsequent projection, distance, and angle calculations, [the following is omitted as it is not part of the original text]. Normalization yields the unit direction vector of the initial axis of the anchor bolt. : (8), among which, For the first The principal direction vector of the anchor bolt. Let be the normalized unit direction vector of the initial axis of the anchor bolt, i.e., the initial axis unit direction vector, and .

[0048] Therefore, we can obtain the first... Initial axis reference point of anchor bolt unit direction vector of the initial axis This process can quickly pinpoint the approximate orientation and location of the anchor bolts, providing a baseline for subsequent steps. It also separates the target anchor bolt cluster from adjacent walls and protruding interference structures, providing an object basis for subsequent main bolt extraction and exposed parameter measurement.

[0049] First, the point cloud sub-data is collected along the unit direction of the initial axis of the anchor bolt. Axial analysis was performed to identify the thick cluster area near the root, the tray area, and the wall transition area, and point cloud sub-data with off-axis scales significantly larger than the main body scale of the thin rod were removed. Subsequently, from the retained point cloud sub-data, data with relatively close off-axis location, high local linearity, and strong directional consistency were extracted first to form the second... The cloud sub-data of the main center point of the anchor bolt is denoted as The system identifies and removes points with off-axis anomalies such as thick clusters at the root and trays along the initial axis direction, while prioritizing the retention of rod points that are close to the axis and have high linearity. This approach can specifically remove interference from the non-thin rod main structure at the anchor root, forming cloud-like data of the main trunk center point that better represents the rod's orientation, thus avoiding the dominance of local abnormal structures in axis determination.

[0050] Furthermore, based on the cloud sub-data of the backbone center point Based on this foundation, axis refinement is performed, and if necessary, continuous point filling and trunk reconstruction are carried out along the axis for short-distance fracture zones to obtain enhanced trunk center point cloud data. Then based on Perform robust line fitting again to obtain the first... Target axis reference point of anchor bolt and the unit direction vector of the target axis Enhancing the continuity and representativeness of the anchor bolt's main structure provides a more reliable foundation for the axis and main structure point cloud, enabling subsequent identification of stable exposed sections and adaptive determination of exposed starting points. Axial point supplementation and fracture zone reconstruction improve the continuity of the main structure's center point cloud sub-data. Subsequent secondary fitting based on the purified, high-quality point cloud sub-data significantly improves the stability and accuracy of the target axis, ensuring the accuracy of the target axis reference point and the target axis unit direction vector, effectively resisting local interference in complex scenarios.

[0051] To improve the stability of axis fitting, after obtaining the candidate point cloud of the anchor bolt trunk and the initial fitted axis, outliers that significantly deviate from the axis can be eliminated based on the distance of each point to the initial fitted axis. Then, iterative fitting is performed again based on the retained points to reduce the impact of local noise, attachment points, or mis-segmented points on the axis direction estimation. For example... Figure 5 As shown, Figure 5 This is a diagram showing the refined anchor bolt axis of an embodiment of this application. Figure 5In the diagram, the fine axis represents the final axis direction obtained after outlier removal and iterative fitting of the single anchor bolt point cloud, while the wall normal represents the normal direction of the tunnel wall or local cross-section reference surface. The auxiliary display of the fine axis and wall normal allows for a direct reflection of the spatial relationship between the anchor bolt trunk direction and the tunnel support surface, and verifies the consistency between the fitted anchor bolt axis and the anchor bolt trunk point cloud distribution. Since the measured 3D point cloud of underground tunnels may contain tray structures, local thick masses, wall transition zones, attachments, and mis-segmented background points near the anchor bolt roots, further trunk structure enhancement extraction and axis refinement after initial axis determination help reduce the interference of non-anchor bolt main structures on subsequent measurements of exposed length and deflection angle.

[0052] In some embodiments, determining the exposed start and end points of the anchor bolt based on the point cloud sub-data, the target axis reference point, and the target axis unit direction vector includes: projecting the point cloud sub-data along the target axis unit direction vector to obtain a one-dimensional projection sequence containing multiple projection values; discretizing the one-dimensional projection sequence to obtain an interval sequence containing multiple axial intervals; determining the corresponding weighted point density of each axial interval in the interval sequence based on the target axis reference point and the target axis unit direction vector; performing an exposure judgment on each axial interval in the interval sequence based on the corresponding weighted point density of each axial interval to obtain at least one candidate exposure interval; and for each... Candidate exposed intervals are identified, and local axis reference points corresponding to these candidate exposed intervals are determined. Based on the target axis unit direction vector and the local axis reference points, the average direction deviation angle of the candidate exposed intervals relative to the target axis unit direction vector is determined. Based on the average direction deviation angle corresponding to each candidate exposed interval, at least one exposed interval is selected from all candidate exposed intervals. Based on the target axis reference point, the target axis unit direction vector, the weighted point density corresponding to each axial interval, and the average direction deviation angle corresponding to each candidate exposed interval, the exposed starting point of the anchor bolt is determined. The maximum projection value is found among all exposed intervals, and the maximum projection value is determined as the exposed ending point corresponding to the anchor bolt.

[0053] In this embodiment, the point cloud sub-data is projected along the unit direction vector of the target axis to obtain a one-dimensional projection sequence containing multiple projection values, so as to convert the point cloud distribution in the original three-dimensional space into a one-dimensional sequence representation along the anchor axis direction.

[0054] For the Any data in the root anchor point cloud sub-data The projection value is determined using the following formula: (9) in, For point The axial projection scalar along the anchor bolt axis, i.e., the projection value. This represents the vector dot product operation.

[0055] Through the above projection, the first The root anchor point cloud is represented as a set of one-dimensional projection sequences according to the axis coordinates. : Compared to directly determining the root and end points in three-dimensional space, this one-dimensional sequence representation can more directly reflect the distribution continuity, density changes, and local missingness of the anchor point cloud along the axial direction, which is beneficial for subsequent identification of exposed regions.

[0056] Furthermore, the one-dimensional projection sequence is discretized to obtain an interval sequence containing multiple axial intervals. Each axial interval includes multiple projection values. The one-dimensional projection sequence is then divided into intervals of fixed length. Discretely divided into several axial intervals: (10) in, For the j-th axial interval of the i-th anchor rod, Let j be the starting point of the j-th axial interval. Let j be the endpoint of the j-th axial interval. The length of the axial discrete interval, its value range is: , This represents the total number of axial intervals.

[0057] By dividing the anchor bolt point cloud into axial bins, the continuity, density variation, and scale variation along the axial direction can be transformed into an analyzable interval sequence, providing a basis for subsequent identification of exposed intervals.

[0058] To reduce the impact of off-axis noise points, attachment points, and local missegmentation points on the interval statistical results, this application introduces a weighted point density analysis method based on the distance from a point to the axis, on top of the one-dimensional axial interval, to preliminarily identify whether each axial interval belongs to the effective exposed section of the anchor bolt. Based on the target axis reference point and the target axis unit direction vector, the corresponding weighted point density of each axial interval in the interval sequence is determined. Specifically, determining the vertical distance of the anchor bolt based on the target axis reference point and the target axis unit direction vector accurately reflects the degree of deviation of each point from the main body of the bolt, which is a prerequisite for suppressing off-axis interference points. For the first... Any point in the anchor point cloud set Define its vertical distance to the anchor bolt axis as: (11), where, due to The unit direction vector of the target axis is given by the above formula, therefore the point is directly represented by the above formula. The perpendicular distance to the line.

[0059] For each axial interval, a distance-defined weight value is determined based on the vertical distance and the corresponding projection value of the axial interval. An exponential decay method is used to define the point weight value, with the weight monotonically decreasing as the distance from the point to the axis increases. This effectively reduces the weight proportion of off-axis noise, attachments, and mis-segmented points, allowing points closer to the main trunk to dominate subsequent density calculations and weakening the interference from non-trunk structures. For the axial interval... Points within the axis are weighted according to their distance from the axis. The weight value is related to The monotonically decreasing function adopts an exponential decay form: (12) in, Here is the weight decay coefficient, and This is used to control the rate at which the weight decreases after a point deviates from the axis; for example, its value range is... .

[0060] Based on the distance-defined weight values ​​corresponding to the axial intervals, the weighted point density of the axial intervals is determined, realistically reflecting the distribution of rod points within each axial interval, and providing a more realistic density index for subsequent interval continuity judgments. Therefore, the axial intervals... The weighted point density is defined as: (13), among which, The length of the axial interval. This represents the weighted point density per unit axial length.

[0061] To reduce the impact of off-axis noise points, attachment points, and local missegmentation points on the interval statistical results, this application, based on a one-dimensional axial interval, performs a preliminary identification of whether each axial interval belongs to the effective exposed section of the anchor bolt. Based on the corresponding weighted point density of each axial interval, an exposure judgment is made for each axial interval in the interval sequence, resulting in at least one candidate exposed interval. Introducing weighted point density analysis can effectively reduce the interference of off-axis noise points, attachment points, and local missegmentation points on the interval statistics. By relying on the one-dimensional axial interval to complete the preliminary screening of the effective exposed section of the anchor bolt, interfering intervals are eliminated in advance, the analysis scope is narrowed, and the bias caused by noise is avoided when performing complex statistics directly on the entire interval, making the exposed interval judgment results more closely reflect the actual shape.

[0062] Density variation alone can still be affected by localized attachment noise or isolated missegmentation points; therefore, this application further introduces a directional consistency criterion. For candidate exposed regions... For points within the anchor bolt, calculate the angle between them and the unit direction vector relative to the target axis of the anchor bolt. Candidate exposed area. The local axis reference point of the inner point cloud is determined by the following formula: (14) in, Candidate exposed interval The number of interior points, Let p be any point cloud within the candidate exposed interval, and let p be the local axis reference point. By averaging all points within the candidate exposed interval, the local axis reference point is obtained, which provides the geometric center of the point cloud distribution in that interval. This provides a unified and reliable local benchmark for subsequent calculations of the directional deviation angle between the points and the axis, ensuring the accuracy of the deviation angle calculation.

[0063] Based on the target axis unit direction vector and the local axis reference point, the direction deviation angle corresponding to each data point in the candidate exposure interval is determined. For any point in the candidate exposure interval... The directional deviation angle relative to the anchor bolt axis is defined as: (15) The use of absolute values ​​in the formula is to eliminate angular ambiguity caused by the opposing orientations of the anchor bolt axis, ensuring that the included angle always represents the degree of deviation of the point distribution from the anchor bolt axis. By calculating the directional deviation angle of each point relative to the target axis, the degree of deviation of the distribution direction of each point from the actual direction of the anchor bolt can be quantified, providing basic data for subsequent interval direction consistency judgment.

[0064] Based on the orientation deviation angle corresponding to each data point in the candidate exposure interval, the average orientation deviation angle of the candidate exposure interval relative to the unit orientation vector of the target axis is determined, and the interval is further defined. The average directional deviation angle is: (16) The average directional deviation angle can comprehensively reflect the overall deviation of the point cloud distribution of the entire interval from the anchor axis, avoid the influence of a single abnormal point on the direction judgment, and provide a stable and reliable overall directional consistency index for subsequent interval screening.

[0065] Based on the average directional deviation angle corresponding to each candidate exposed section, at least one exposed section is selected from all candidate exposed sections. The point cloud distribution of the actual exposed section of the anchor bolt closely matches the direction of the bolt axis and has a small directional deviation angle, while noise, attachments, and mis-segmented areas show obvious angular deviations. This can further eliminate invalid sections remaining from the initial screening, compensate for the shortcomings of a single density criterion, significantly improve the accuracy of exposed section identification, and make the judgment results more consistent with the actual shape of the anchor bolt.

[0066] Based on the target axis reference point, the target axis unit direction vector, the corresponding weighted point density for each axial interval, and the average direction deviation angle for each candidate exposed interval, the exposed starting point of the anchor bolt is determined. This method can comprehensively judge by combining multiple information such as the axis reference point, unit direction vector, weighted point density, and average direction deviation angle, thus overcoming the limitations of single data judgment. It can accurately distinguish between the wall-fitting area and the exposed area of ​​the bolt, effectively avoiding position misjudgment caused by debris on the tunnel wall and point cloud noise, making the location of the exposed starting point more consistent with the actual situation on site, and improving the positioning accuracy and reliability.

[0067] The maximum projection value is found among all exposed intervals and is represented as follows: (17), among which, For the maximum projection value, Let be the scalar value obtained by projecting point P along the unit direction vector of the target axis. It represents the position of the point on the axis, i.e., the projected value. Let be the set of all exposed sections of the i-th anchor bolt. The maximum projection value is determined as the exposed endpoint corresponding to the anchor bolt.

[0068] In addition, the coordinate representation of the exposed endpoint in three-dimensional space can be determined by the following formula: (18), among which, The coordinates of the exposed endpoint in three-dimensional space are used. Based on the actual shape characteristics of the anchor rod extending along the axis, the maximum axial projection value within the exposed section is selected as the exposed endpoint. This allows for precise positioning of the outermost end of the anchor rod away from the wall, ensuring the judgment logic aligns with the actual structure. Furthermore, by taking values ​​only within the selected valid exposed section, interference from scattered noise points and attachments at the end is avoided, preventing endpoint position deviation and ensuring accurate and stable positioning results. This also lays a reliable foundation for subsequent accurate calculation of the exposed length.

[0069] In some embodiments, the step of determining the exposure of each axial interval in the interval sequence based on the corresponding weighted point density of each axial interval to obtain at least one candidate exposed interval includes: determining the change in weighted point density of each pair of adjacent axial intervals in the interval sequence based on the weighted point density of each axial interval; determining a density change threshold based on the weighted point density of all axial intervals; and for each pair of adjacent axial intervals, in response to determining that the change in weighted point density is less than or equal to the density change threshold, designating each axial interval corresponding to the pair of adjacent axial intervals as a candidate exposed interval.

[0070] In this embodiment, based on the weighted point density corresponding to each axial interval, the change in weighted point density between each pair of adjacent axial intervals in the interval sequence is determined. Calculating the change in weighted point density between adjacent axial intervals can capture the abrupt density changes between the pole and non-pole areas such as walls and trays, providing a direct basis for subsequent threshold screening of continuous pole intervals and facilitating the identification of continuous pole distribution segments. The change in weighted point density between adjacent axial intervals is calculated using the following formula: (19), among which, The weighted point density for the (j+1)th axial interval. Let be the weighted point density of the j-th axial interval.

[0071] A density change threshold is determined based on the weighted point density corresponding to all axial intervals. Specifically, the average weighted point density of the anchor bolts is determined based on the weighted point density corresponding to all axial intervals. A density change threshold is then determined based on the average weighted point density of the anchor bolts. The judgment criteria are adaptively set according to the overall density level of the anchor bolts themselves to avoid misjudgments under different point cloud density scenarios with a fixed threshold, ensuring the rationality of density change judgment. To improve the adaptability of the method under different point cloud density conditions, the average weighted point density of the anchor bolts along the axial direction is defined as: (20), among which, This represents the total number of axial intervals.

[0072] The density change threshold is determined using the following formula: (21), among which, It is an empirical coefficient, and For example, its value range is .

[0073] For each pair of adjacent axial intervals, if the density change at the weighted point is less than or equal to the density change threshold, each axial interval corresponding to the pair of adjacent axial intervals is considered a candidate exposed interval. That is, when the density change at the weighted point corresponding to a pair of adjacent axial intervals satisfies the following formula: (22) Then each axial interval corresponding to the two adjacent axial intervals is taken as a candidate exposed interval. Taking the adjacent axial intervals with the weighted point density change less than or equal to the density change threshold as candidate exposed intervals can effectively screen out intervals with gentle density changes that conform to the continuous distribution characteristics of the pole, and eliminate interference intervals formed by noise, attachments and missegmentation points, so as to provide a reliable candidate range for the accurate identification of the exposed section in the future.

[0074] In some embodiments, the step of selecting at least one exposed interval from all candidate exposed intervals based on the average directional deviation angle corresponding to each candidate exposed interval includes: for each candidate exposed interval, in response to determining that the average directional deviation angle corresponding to the candidate exposed interval is less than or equal to a directional consistency threshold, determining the candidate exposed interval as the exposed interval.

[0075] In this embodiment, the following condition is met when the average direction deviation angle is less than or equal to the direction consistency threshold: (23) It is believed that the point cloud in the candidate exposed area has a high degree of consistency with the direction of the anchor bolt axis, among which, This is a directional consistency threshold, for example, with a value range of... When the point cloud quality is high and the anchor points are relatively concentrated, Take the smaller value; when the point cloud exhibits local discretization. The value is appropriately increased. If the average directional deviation angle is less than or equal to the directional consistency threshold, the candidate exposed region is considered to meet the directional consistency requirement, effectively eliminating interference regions where the density meets the requirements but the point cloud distribution direction deviates significantly from the pole's orientation. However, selecting candidate exposed regions solely based on density variation and directional consistency may still be affected by localized attachment noise or isolated missegmentation points. Therefore, this application further incorporates off-axis scale stability for a comprehensive judgment. Figure 6 As shown, Figure 6 This is a multi-index curve diagram of the stability of the exposed section in this application embodiment. It is used to show the changing trends of point density, directional consistency and off-axis dimensional stability of each axial section after the anchor bolt is divided into boxes along the axial direction, and to provide a visual basis for the subsequent determination of continuous and stable exposed sections.

[0076] In some embodiments, determining the starting point for the anchor bolt to emerge from the wall based on the target axis reference point, the target axis unit direction vector, the corresponding weighted point density for each axial interval, and the average direction deviation angle for each candidate exposed interval includes: for each axial interval, determining the weighted average off-axis scale corresponding to the axial interval based on the target axis reference point and the target axis unit direction vector; finding the weighted average off-axis scale corresponding to each exposed interval among the weighted average off-axis scales corresponding to all axial intervals, and calculating the mean of the weighted average off-axis scales corresponding to all exposed intervals; and determining the stable rod-shaped reference scale based on the mean. In the weighted point density corresponding to each axial interval, the weighted point density corresponding to each exposed interval is found, and in the average direction deviation angle corresponding to each candidate exposed interval, the average direction deviation angle corresponding to each exposed interval is found; for each exposed interval, based on the weighted point density corresponding to the exposed interval, the average direction deviation angle corresponding to the exposed interval, and the stable rod-shaped reference scale, the stability score corresponding to the exposed interval is determined; in response to determining that the stability score is greater than or equal to the interval score threshold, the exposed interval is taken as the target axial interval; the minimum projection value is found in all target axial intervals, and the minimum projection value is determined as the exposed starting point.

[0077] In this embodiment, this application does not use the geometric intersection of the anchor bolt axis and the roadway wall plane as the sole exposed starting point. Instead, it utilizes the continuous distribution pattern of the anchor bolt point cloud near the root along the axial direction, the dimensional stability of its own rod-shaped cross-section, and directional consistency to directly and adaptively determine the exposed starting point. Figure 7 As shown, Figure 7 A parameter curve is used to determine the exposed starting point in the embodiments of this application, wherein, Figure 7 The median off-axis scale within an interval is the median statistical measure of the perpendicular distance from each point within the corresponding axial interval to the anchor axis, used to characterize the typical off-axis scale of the point cloud relative to the anchor axis within that interval. The number of points in the axial interval characterizes the number of points participating in the statistics within the corresponding axial interval, reflecting the continuity of the point cloud distribution and the degree of data support within that interval. The technical essence of these two aspects lies in off-axis scale statistics and axial binning point count statistics, respectively. Figure 7It is evident that stable slender rod regions typically exhibit a midpoint off-axis scale close to the stable rod-like reference scale with relatively gentle changes between adjacent intervals, while maintaining a certain continuity in the number of points in the axial intervals. Conversely, thick root regions, tray regions, or wall transition regions tend to show larger or abrupt off-axis scales, and abnormal increases or decreases in the number of points in the axial intervals. Therefore, the midpoint off-axis scale and the number of points in the axial intervals can serve as visual interpretations and auxiliary verification parameters for the stability score determination results, explaining why the smallest axial projection value that meets the stability score threshold and is located at the beginning of a continuous stable interval is selected as the final exposed starting point. It should be noted that this application does not first determine an initial exposed starting point and then make corrections, but rather directly determines the final exposed starting point based on the aforementioned criteria of continuity, scale stability, and directional consistency.

[0078] Let the first The target axis reference point for the anchor bolt is The unit direction vector of the target axis is Then its axis parametric equation is expressed as: (24), among which, For axis parameters.

[0079] For the Any point in the root anchor point cloud sub-data Its axial projection scalar is denoted as: (25) Based on the obtained axial discrete interval, the first Each axial interval is denoted as For the axial interval For each point within the axis, using the defined perpendicular distances from the points to the axis and their weights, the weighted average off-axis scale for this axial interval is further defined as follows: (26), among which, For distance decay weights, Point The vertical distance to the anchor bolt axis. (27). By combining distance attenuation weight to weaken the influence of abnormal points that deviate from the center of the rod, the actual radial thickness of the anchor rod in each segment can be accurately reflected, and the cross-sectional shape characteristics of the anchor rod itself can be accurately reflected.

[0080] Because the anchor bolt point cloud corresponding to the exposed stable section has a relatively stable slender rod-like shape in the axial direction, its average off-axis scale in the axial range is... The variation is small within continuous intervals. Based on this, among the weighted average off-axis scales corresponding to all axial intervals, the weighted average off-axis scale corresponding to each exposed interval is found, and the mean of the weighted average off-axis scales corresponding to all exposed intervals is calculated. Based on the mean, a stable rod-shaped reference scale is determined, defined as the _th_ _ The stabilizing rod-shaped reference dimensions of the anchor bolt: (28), among which, Indicates the first The set of intervals where the anchor bolts meet the exposure criteria. The weighted average off-axis dimension of all exposed intervals is calculated as the reference dimension for stable rods. This allows us to extract the standard thickness of the normally exposed rod body, which serves as a unified evaluation benchmark to accurately distinguish between stable rod areas and abnormally sized root areas or areas affected by debris.

[0081] Within the weighted point density corresponding to each axial interval, the weighted point density corresponding to each exposed interval is found. Similarly, within the average directional deviation angle corresponding to each candidate exposed interval, the average directional deviation angle corresponding to each exposed interval is found. Multiple axial intervals contain multiple candidate exposed intervals, and these candidate exposed intervals further contain multiple exposed intervals. Therefore, the weighted point density corresponding to each exposed interval can be found within the weighted point density corresponding to multiple axial intervals, and the average directional deviation angle corresponding to each exposed interval can be found within the average directional deviation angle corresponding to multiple candidate exposed intervals. For each exposed interval, based on its corresponding weighted point density, average directional deviation angle, and stable rod-shaped reference scale, a stability score is determined. By comprehensively calculating the stability score using weighted point density, average directional deviation angle, and stable rod-shaped reference scale, exposed intervals can be evaluated from multiple dimensions, including point cloud density, distribution direction, and morphological characteristics. This comprehensively identifies effective exposed areas, avoids the limitations of single-indicator judgments, effectively eliminates interference from noise, attachments, and other factors, and improves the accuracy and stability of interval determination.

[0082] In response to determining that the stability score is greater than or equal to the interval score threshold, the exposed interval is taken as the target axial interval. Let the continuous stability interval length threshold be... The interval score threshold is If a certain exposed interval sequence satisfies: (29), among which, For stability scoring, if in a certain interval sequence, starting from the th... From the exposed area onwards The stability scores of each exposed interval are all greater than or equal to the interval score threshold, and the first interval... The exposed section is the exposed section closest to the anchor root that satisfies the above continuous stability condition. Therefore, the first exposed section is... The exposed section is defined as the starting section of the stable exposed section, and the anchor bolt is considered to have entered the stable exposed section from this section. This allows for precise screening of stable exposed sections of anchor bolts that meet the standards in terms of shape, density, and direction, eliminating root distortion sections and disturbed abnormal sections, and locking in a reliable exposed range.

[0083] Therefore, the minimum projection value is found in the entire target axial interval, and the minimum projection value is determined as the exposed starting point. The minimum interval number that satisfies the above conditions is then selected. Then the first The axial scalar value of the exposed starting point of the anchor bolt is defined as: (30), among which, Indicates the first The minimum projection value of each axial interval.

[0084] Furthermore, the coordinates of the exposed starting point in three-dimensional space can be represented as: , (31).

[0085] The aforementioned method for determining the starting point essentially abandons the geometric intersection method. Instead, it relies on the axial continuity, directional consistency, and rod-shaped dimensional consistency of the anchor bolt point cloud itself to determine the exposed starting point. It does not depend on local wall plane fitting or line-plane intersection calculations, making it more suitable for real-world testing scenarios in underground tunnels with rough walls, localized attachment interference, and incomplete root point clouds. Selecting the minimum axial projection value within a suitable stable exposed range as the exposed starting point accurately pinpoints the initial position where the anchor bolt just detaches from the wall and officially begins to be exposed, closely matching the actual anchor bolt layout. By utilizing effective range values ​​that have eliminated interference, it avoids misjudgments caused by wall irregularities, root obstruction, and anomalies. The starting point determination standard is unified and calculation is simple, effectively improving the accuracy of the exposed starting point positioning and ensuring the reliability of subsequent calculations of exposed length and radial deflection angle.

[0086] In some embodiments, determining the stability score corresponding to the exposed interval based on the weighted point density corresponding to the exposed interval, the average directional deviation angle corresponding to the exposed interval, and the stable rod-shaped reference scale includes: normalizing the weighted point density corresponding to the exposed interval to obtain a normalized weighted point density value; determining the average directional deviation angle score based on the average directional deviation angle corresponding to the exposed interval; determining the off-axis scale consistency score based on the weighted average off-axis scale and the stable rod-shaped reference scale corresponding to the exposed interval; and determining the stability score based on the weighted point density normalized value, the average directional deviation angle score, and the off-axis scale consistency score.

[0087] In this embodiment, to facilitate the fusion of point density, directional consistency, and off-axis scale stability under the same dimension, this application further converts the weighted point density, average directional deviation angle, and weighted average off-axis scale corresponding to each exposed interval into normalized evaluation indicators. Normalizing the weighted point density corresponding to the exposed interval yields a normalized value, unifying the data dimensions and enabling the fusion calculation of different indicators. Based on the average directional deviation angle corresponding to the exposed interval, an average directional deviation angle score is determined. Specifically, the sum of the directional normalization reference threshold and a small positive number is calculated, and then a first ratio of the average directional deviation angle corresponding to the exposed interval to the above sum is calculated. The average directional deviation angle score is determined based on the result after subtracting this first ratio. The smaller the average directional deviation angle, the more consistent the point cloud distribution direction is with the anchor axis direction, resulting in a higher score. By determining the average directional deviation angle score, the degree of fit between the point cloud direction and the anchor axis can be quantified, intuitively reflecting directional consistency. Based on the weighted average off-axis scale and stable rod-shaped reference scale corresponding to the exposed interval, the off-axis scale consistency score is determined. Specifically, the absolute value of the difference between the weighted average off-axis scale and the stable rod-shaped reference scale is calculated. This absolute value is then divided by the sum of the scale consistency adjustment parameter and a small positive number to obtain a second ratio. The negative exponent of this second ratio is then used to obtain the off-axis scale consistency score. According to the... The off-axis scale consistency score is determined by comparing the weighted average off-axis scale of each exposed section with the stable rod-shaped reference scale. The closer the weighted average off-axis scale is to the stable rod-shaped reference scale, the more the section conforms to the dimensional characteristics of a stable slender rod segment, and the higher the corresponding score. By determining the off-axis scale consistency score, it is possible to assess whether the section scale conforms to the characteristics of the anchor bolt body; the closer the scale is to the standard shape, the higher the score.

[0088] thus, Figure 6 The average directional deviation angle score and off-axis scale consistency score correspond to directional consistency evaluation and rod-shaped scale consistency evaluation, respectively. They are not new technical concepts independent of the aforementioned average directional deviation angle, weighted average off-axis scale, and stable rod-shaped reference scale.

[0089] Based on the above weighted point density normalized value, average direction deviation angle score, and off-axis scale consistency score, an axial interval stability score function is further constructed to determine the stability score corresponding to the exposed interval: , (32), In formula (32) The overall representation is the weighted point density normalized value. The overall score represents the average directional deviation angle. Overall, the score represents the consistency of off-axis scale. For the first Weighted point density of each axial interval For the first Axial average weighted density of anchor bolts This is the average directional deviation angle for this axial interval. As a reference threshold for direction normalization, This is a parameter for adjusting scale consistency. To prevent tiny positive numbers with a denominator of zero; Let the weighting coefficients satisfy: (33) Wherein, when it is desired to enhance the constraint on the stability of the rod-shaped section, the appropriate increase is made. The value of should be appropriately increased when the point cloud density fluctuates significantly. The values ​​of [values ​​to be determined] are used to construct an axial interval stability scoring function that integrates density, directional deviation, and rod size. This function comprehensively evaluates the interval state from multiple dimensions and allows for flexible adjustment of weighting coefficients to adapt to different point cloud conditions. It considers both the density of point cloud distribution and strengthens the constraints of rod shape, resulting in a more comprehensive and reasonable evaluation. Finally, the stability score is obtained by fusing the three types of scores. This comprehensive evaluation of the exposed interval from multiple dimensions, including density distribution, directional deviation, and shape scale, effectively avoids the shortcomings of a single indicator and improves the reliability of the results.

[0090] Combination Figure 6 It is evident that the weighted point density normalized value, average orientation deviation angle score, off-axis scale consistency score, and stability score exhibit phased changes along the axial projection position. Among these, stable exposed regions are typically characterized by a consistently high stability score within adjacent axial intervals, with no drastic abrupt changes in point density, orientation consistency, or off-axis scale stability-related indicators. Conversely, regions with low stability scores, or those showing significant drops or large fluctuations within adjacent axial intervals, usually correspond to thick root clusters, attached noise, local missegmentation points, or missing point clouds, and are unsuitable as stable exposed regions. Therefore, within exposed regions that already meet the density change and orientation consistency criteria, determining continuous intervals that further meet the stability score criterion as target axial intervals can improve the accuracy and reliability of exposed region identification.

[0091] In some embodiments, determining the exposed parameters of the anchor bolt based on the target axis reference point, the target axis unit direction vector, the exposed start point, and the exposed end point includes: acquiring the axis data of the roadway, determining the data with the smallest distance to the exposed start point as the nearest point; acquiring other data within a predetermined range from the axis data that are within the nearest point; determining the local roadway axis direction based on the nearest point and the other data; determining the roadway cross-section reference plane based on the local roadway axis direction; projecting the target axis unit direction vector onto the roadway cross-section reference plane to obtain an in-plane projection vector, and normalizing the in-plane projection vector to obtain a unit radial vector; determining the radial deflection angle corresponding to the anchor bolt based on the angle between the target axis unit direction vector and the unit radial vector; determining the exposed length of the anchor bolt based on the exposed start point and the exposed end point; and using the radial deflection angle and the exposed length corresponding to the anchor bolt as the exposed parameters corresponding to the anchor bolt.

[0092] In this embodiment, after determining the exposed starting point, this application further calculates the first... The radial deflection angle and exposed length of the anchor bolt are considered. The radial deflection angle characterizes the degree of deviation of the exposed section axis of the anchor bolt from the radial reference direction within the roadway cross-section. This application determines the roadway cross-section reference plane based on the local roadway axis direction near the exposed anchor bolt starting point, and projects the unit direction vector of the anchor bolt target axis onto this cross-section reference plane to obtain the unit radial vector used for calculating the radial deflection angle. Specifically, roadway axis data is extracted based on three-dimensional point cloud data within the roadway, and the axis point closest to the exposed anchor bolt starting point is determined from the roadway axis data. This closest axis point is used to locate the local reference position of the cross-section where the anchor bolt is located, thus providing a spatial reference for determining the local roadway axis direction and calculating the deflection angle. Other data within a predetermined range from the nearest point are obtained from the axis data. Selecting axis data within the predetermined range of the nearest point eliminates irrelevant points at a distance, focuses the target area, and simultaneously reduces single-point errors and noise effects by utilizing multi-point continuous features, ensuring the accuracy and reliability of the fitted local roadway axis direction. Based on the nearest point and other data, the local tunnel axis direction is determined. By fitting the local tunnel axis direction using the aforementioned points, the true direction of the tunnels in the target area can be accurately represented. Let the direction be... Exposed starting point of anchor bolt The closest point of the corresponding tunnel axis is And let the direction vector of the local tunnel axis at this location be... Local tunnel axis direction vector The reference plane of the roadway cross-section can be used to determine the location of the anchor bolt. The first... Unit direction vector of the target axis of the anchor bolt Project the vector onto the reference plane of the tunnel cross-section and normalize it to obtain the unit radial vector used for calculating the radial deflection angle. : (34) in, Here is the unit radial vector used for calculating the radial deflection angle. For the first The unit direction vector of the target axis of the anchor bolt. This refers to the direction vector of the local roadway axis near the exposed starting point of the anchor bolt. By projecting the unit direction vector of the target axis onto a roadway cross-sectional reference plane perpendicular to the local roadway axis, the radial reference direction within the roadway cross-section can be obtained. This unifies the reference standard for determining the radial deflection angle of the anchor bolt and ensures a unified and standardized angle calculation benchmark.

[0093] Therefore, the first The radial deflection angle of an anchor bolt is defined as follows: (35) because and Both are unit vectors, and the formula uses absolute values. Therefore, this definition can eliminate the angular ambiguity caused by the positive and negative directions of the anchor bolt axis, ensuring that the radial deflection angle always represents the degree of deviation of the anchor bolt axis from the unit radial vector within the roadway cross-section. The radial deflection angle is the angle between the unit direction vector of the anchor bolt target axis and the unit radial vector, used to accurately quantify the degree of deviation of the anchor bolt placement direction from the radial direction of the roadway cross-section.

[0094] like Figure 8 As shown, Figure 8 This is a diagram showing the radial deflection angle measurement results of an embodiment of this application. Figure 8 In the diagram, the anchor bolt axis represents the first... The target axis direction of the exposed section of the anchor bolt, the unit radial vector represents the radial reference direction within the cross-section obtained by projecting the target axis unit direction vector onto the roadway cross-section reference plane, the local roadway axis direction is used to determine the roadway cross-section reference plane where the anchor bolt is located, and the exposed start and end points are used to define the spatial range of the exposed section of the anchor bolt. Figure 8 As can be seen, the radial deflection angle is the spatial angle between the unit direction vector of the anchor bolt target axis and the unit radial vector. This angle is used to quantify the degree of deviation of the exposed section axis of the anchor bolt from the radial reference direction of the roadway cross-section.

[0095] The calculation of the exposed length is based on the exposed start and exposed end points to determine the exposed length of the anchor bolt. For example... Figure 9 As shown, Figure 9 This is a diagram showing the measurement results of the exposed start point, exposed end point, and exposed length in an embodiment of this application. Figure 9In this diagram, the exposed start point indicates the boundary position where the anchor bolt begins to protrude from the surrounding rock or wall surface; the exposed end point indicates the end position of the exposed section of the anchor bolt away from the wall surface; and the final exposed measurement axis is used to indicate the axial measurement direction corresponding to the exposed length. Figure 9 It is evident that the exposed length is not directly calculated using the ordinary Euclidean distance between the two endpoints. Instead, it is determined along the direction of the final exposed measurement axis by the difference between the axial projection scalar corresponding to the exposed starting point and the axial projection scalar corresponding to the exposed ending point. This reduces the length measurement deviation caused by end discrete points, local bending noise, or occlusion.

[0096] Then the first The exposed length of an anchor bolt is defined as follows: (36) To avoid negative values ​​in abnormal situations, it is further written as: (37) The above length definition refers to the effective exposed length of the anchor rod along its own axis, rather than the Euclidean distance between any two points. Therefore, it can reduce the impact of anchor rod end dispersion, local bending noise, and point cloud occlusion on the measurement results. Unlike the method of directly using the Euclidean distance between the two ends, this definition fits the actual extension shape of the anchor rod, effectively reducing the measurement error caused by point cloud dispersion, slight bending, and local point cloud occlusion at the end of the rod, and ensuring accurate and stable measurement of the exposed length.

[0097] like Figure 10 As shown, Figure 10 This is a diagram showing the overall exposed length measurement and radial deflection angle results of an embodiment of this application. Figure 10 In this context, the exposed starting point, exposed ending point, and final measurement axis collectively characterize the overall spatial position of the exposed section of the anchor bolt. The unit radial vector is the normalized direction of the projection of the unit direction vector of the anchor bolt target axis onto the reference plane of the roadway cross-section, used to characterize the cross-sectional radial reference direction upon which the radial deflection angle calculation is based. Figure 10 As can be seen, the exposed length of the anchor bolt can be determined based on the exposed starting point, exposed ending point, and final measurement axis; the radial deflection angle of the anchor bolt can be determined based on the spatial angle between the unit direction vector and the unit radial vector of the final measurement axis. Therefore, using the exposed length and radial deflection angle of the anchor bolt as its corresponding exposed parameters can intuitively reflect the on-site installation posture and actual exposed dimensions of the anchor bolt, and provide quantitative basis for roadway support quality inspection, construction compliance verification, and support status data analysis.

[0098] Furthermore, the measurement results of this method were compared with the actual results on site. The actual results on site showed that the exposed length of the anchor rod was 8.3 cm, while the exposed length measured by this method was 8.18 cm. The two results are close, and the measurement error is small. The radial deflection angle of the anchor rod measured by this method was 1.52°, which is consistent with the actual situation on site, indicating that this method has good measurement accuracy and engineering applicability.

[0099] In another embodiment provided in this application, such as Figure 11 As shown, Figure 11 This is a flowchart illustrating a method for measuring exposed parameters of roadway anchor bolts according to another embodiment of this application. By sequentially performing preprocessing on the 3D point cloud data within the roadway, the following steps are performed: anchor bolt candidate point extraction, single anchor bolt instance separation, initial axis fitting, data augmentation followed by re-axis fitting, axis refinement, axial projection and interval construction, candidate exposed interval identification, exposed endpoint determination, adaptive determination of exposed starting point, and calculation of radial deflection angle and exposed length. This achieves automatic measurement of the exposed length and radial deflection angle of a single anchor bolt under measured point cloud conditions in underground roadways. During implementation, intermediate and final results are simultaneously output, including the target anchor bolt cluster, the main trunk center zone, multi-index curves for the stability of exposed intervals, starting point determination results, starting and ending point measurement results, and radial deflection angle results. These results can be used for debugging analysis, manual verification, and demonstration of implementation effects.

[0100] This application, based on the 3D point cloud of underground tunnels, achieves automatic quantitative measurement of the exposed length and radial deflection angle of anchor bolts through anchor bolt candidate point cloud extraction, single anchor bolt instance separation, main structure enhancement extraction, axis refinement, axial projection and interval construction, candidate exposed interval identification, and adaptive determination of the exposed starting point. Compared to existing technologies that mainly rely on anchor bolt detection or local wall fitting to determine the exposed starting point, this application reduces the dependence of the exposed starting point determination on local wall fitting results and improves the stability and reliability of exposed length measurement under conditions of root obstruction, wall roughness, and attachment interference. By defining the radial deflection angle as the angle between the anchor bolt axis direction and the unit radial vector within the tunnel cross-section, the angle calculation benchmark is unified, enhancing the comparability and engineering interpretability of the measurement results. This application does not rely on a specific scanning platform or a single wall modeling method, and has good adaptability to complex scenes, engineering applicability, and promotional application value.

[0101] It should be noted that the method in this embodiment can be executed by a single device, such as a computer or server. The method can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method in this embodiment, and the multiple devices will interact with each other to complete the method described.

[0102] It should be noted that some embodiments of this application have been described above. In some cases, the actions or steps described in the above embodiments can be performed in a different order than that shown in the above embodiments and the desired result can still be achieved. In addition, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0103] Based on the same inventive concept, corresponding to any of the above embodiments, this application also provides a measuring device for the exposed parameters of roadway anchor bolts.

[0104] refer to Figure 12 The measuring device for the exposed parameters of the roadway anchor bolts includes: The acquisition module 10 is configured to acquire three-dimensional point cloud data in the tunnel and, based on the three-dimensional point cloud data, determine the point cloud sub-data corresponding to each anchor bolt.

[0105] The extraction module 20 is configured to perform axis enhancement extraction on the point cloud sub-data corresponding to each anchor rod to obtain the target axis reference point and the target axis unit direction vector corresponding to the anchor rod.

[0106] The first determining module 30 is configured to determine the exposed starting point and exposed ending point of the anchor bolt based on the point cloud sub-data, the target axis reference point, and the target axis unit direction vector.

[0107] The second determining module 40 is configured to determine the exposed parameters corresponding to the anchor bolt based on the target axis reference point, the target axis unit direction vector, the exposed start point, and the exposed end point.

[0108] The aforementioned device acquires three-dimensional point cloud data within the tunnel. Based on this data, it determines the point cloud sub-data corresponding to each anchor bolt, reducing interference from irrelevant data and accurately distinguishing individual anchor bolts, thus avoiding confusion among multiple anchor bolt data. For each anchor bolt's corresponding point cloud sub-data, axis enhancement extraction is performed to obtain the target axis reference point and target axis unit direction vector. By performing axis enhancement extraction on a subset of the point cloud data for a single anchor bolt, it is possible to filter out cluttered noise and distorted points, accurately pinpointing the anchor bolt's center direction. This not only clarifies the anchor bolt axis reference point and direction vector but also mitigates the influence of rock wall attachments and point cloud incompleteness, accurately restoring the actual anchor bolt posture. Based on the point cloud sub-data, the target axis reference point, and the target axis unit direction vector, the exposed start and end points of the anchor bolt are determined. By relying on the target axis reference point and the target axis unit direction vector to constrain the overall orientation, and without depending on local scattered point cloud morphology, even with obstructions or missing points in the local point cloud, it can still complete interval division and feature discrimination according to the inherent axial pattern of the anchor rod. It avoids misjudging boundary positions due to local data anomalies, thus stably locking the standard wall intersection point, effectively reducing judgment errors and improving the accuracy of exposed start and end point identification. Based on the target axis reference point, the target axis unit direction vector, the exposed start point, and the exposed end point, the corresponding exposed parameters of the anchor rod are determined, effectively reducing position determination and angle calculation errors and significantly improving the accuracy of exposed parameters.

[0109] In some embodiments, the acquisition module 10 is further configured to preprocess the three-dimensional point cloud data, classify the preprocessed three-dimensional point cloud data using a pre-trained point cloud semantic segmentation model to obtain anchor bolt type point cloud data, and cluster the anchor bolt type point cloud data to obtain point cloud sub-data corresponding to each anchor bolt.

[0110] In some embodiments, the extraction module 20 is further configured to perform axis fitting on the point cloud sub-data to obtain an initial axis reference point and an initial axis unit direction vector; perform axial analysis and extraction on the point cloud sub-data along the initial axis unit direction vector to obtain the main trunk center point cloud sub-data corresponding to the anchor bolt; perform data augmentation on the main trunk center point cloud sub-data; and perform axis fitting on the data-augmented main trunk center point cloud sub-data to obtain the target axis reference point and the target axis unit direction vector.

[0111] In some embodiments, the first determining module 30 is further configured to: project the point cloud sub-data along the unit direction vector of the target axis to obtain a one-dimensional projection sequence containing multiple projection values; discretize the one-dimensional projection sequence to obtain an interval sequence containing multiple axial intervals; determine the corresponding weighted point density of each axial interval in the interval sequence based on the target axis reference point and the target axis unit direction vector; perform an exposure judgment on each axial interval in the interval sequence based on the corresponding weighted point density of each axial interval to obtain at least one candidate exposure interval; and for each candidate exposure interval, determine the local area corresponding to the candidate exposure interval. The local axis reference point is used; based on the target axis unit direction vector and the local axis reference point, the average direction deviation angle of the candidate exposed interval relative to the target axis unit direction vector is determined; based on the average direction deviation angle corresponding to each candidate exposed interval, at least one exposed interval is selected from all candidate exposed intervals; based on the target axis reference point, the target axis unit direction vector, the weighted point density corresponding to each axial interval, and the average direction deviation angle corresponding to each candidate exposed interval, the exposed starting point of the anchor rod is determined; the maximum projection value is found in all exposed intervals, and the maximum projection value is determined as the exposed ending point corresponding to the anchor rod.

[0112] In some embodiments, the first determining module 30 is further configured to determine the change in weighted point density corresponding to each pair of adjacent axial intervals in the interval sequence based on the weighted point density corresponding to each axial interval; determine a density change threshold based on the weighted point density corresponding to all axial intervals; and for each pair of adjacent axial intervals, in response to determining that the change in weighted point density is less than or equal to the density change threshold, designate each axial interval corresponding to the pair of adjacent axial intervals as a candidate exposed interval.

[0113] In some embodiments, the first determining module 30 is further configured to, for each candidate exposed interval, determine the candidate exposed interval as the exposed interval in response to determining that the average directional deviation angle corresponding to the candidate exposed interval is less than or equal to the directional consistency threshold.

[0114] In some embodiments, the first determining module 30 is further configured to, for each axial interval, determine the weighted average off-axis scale corresponding to the axial interval based on the target axis reference point and the target axis unit direction vector; find the weighted average off-axis scale corresponding to each exposed interval among the weighted average off-axis scales corresponding to all axial intervals, and calculate the mean of the weighted average off-axis scales corresponding to all exposed intervals; determine a stable rod-shaped reference scale based on the mean; find the weighted point density corresponding to each exposed interval among the weighted point densities corresponding to each axial interval, and find the average direction deviation angle corresponding to each exposed interval among the average direction deviation angles corresponding to each candidate exposed interval; for each exposed interval, determine the stability score corresponding to the exposed interval based on the weighted point density corresponding to the exposed interval, the average direction deviation angle corresponding to the exposed interval, and the stable rod-shaped reference scale; in response to determining that the stability score is greater than or equal to the interval score threshold, designate the exposed interval as the target axial interval; find the minimum projection value among all target axial intervals, and determine the minimum projection value as the exposed starting point.

[0115] In some embodiments, the first determining module 30 is further configured to normalize the weighted point density corresponding to the exposed interval to obtain a weighted point density normalized value; determine an average direction deviation angle score based on the average direction deviation angle corresponding to the exposed interval; determine an off-axis scale consistency score based on the weighted average off-axis scale and the stable rod-shaped reference scale corresponding to the exposed interval; and determine the stability score based on the weighted point density normalized value, the average direction deviation angle score, and the off-axis scale consistency score.

[0116] In some embodiments, the second determining module 40 is further configured to: acquire the axis data of the roadway; determine the data in the axis data that has the smallest distance to the exposed starting point as the nearest point; acquire other data in the axis data that are within a predetermined range of the nearest point; determine the local roadway axis direction based on the nearest point and the other data; determine the roadway cross-section reference plane based on the local roadway axis direction; project the target axis unit direction vector onto the roadway cross-section reference plane to obtain an in-plane projection vector, and normalize the in-plane projection vector to obtain a unit radial vector; determine the radial deflection angle corresponding to the anchor bolt based on the angle between the target axis unit direction vector and the unit radial vector; determine the exposed length of the anchor bolt based on the exposed starting point and the exposed ending point; and use the radial deflection angle and the exposed length corresponding to the anchor bolt as the exposed parameters corresponding to the anchor bolt.

[0117] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, in implementing this application, the functions of each module can be implemented in one or more software and / or hardware.

[0118] The apparatus described above is used to implement the measurement method for the exposed parameters of the roadway anchor bolts in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0119] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the method for measuring the exposed parameters of the roadway anchor bolts as described in any of the above embodiments.

[0120] Figure 13 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0121] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0122] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0123] The input / output interface 1030 is used to connect input / output modules to realize information input and output. The input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0124] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0125] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0126] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0127] The electronic equipment described above is used to implement the measurement method for the exposed parameters of the roadway anchor bolts in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0128] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the method for measuring the exposed parameters of roadway anchor bolts as described in any of the above embodiments.

[0129] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0130] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the method for measuring the exposed parameters of the roadway anchor bolts as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0131] Based on the same concept, corresponding to the methods of any of the above embodiments, this application also provides a computer program product, including computer program instructions, which, when run on a computer, cause the computer to execute the method for measuring the exposed parameters of roadway anchor bolts as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0132] It is understood that before using the technical solutions of the various embodiments in this disclosure, users will be informed of the type, scope of use, and usage scenarios of the personal information involved in an appropriate manner, and user authorization will be obtained.

[0133] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose, based on the prompt message, whether to provide personal information to the software or hardware such as electronic devices, applications, servers, or storage media performing the operations of this disclosed technical solution.

[0134] As an optional but not limited implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0135] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0136] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application is limited to these examples; under the concept of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in detail for the sake of brevity.

[0137] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this application, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this application, and this also takes into account the fact that the details of the implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this application will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this application, it will be apparent to those skilled in the art that the embodiments of this application can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0138] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0139] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of this application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A method for measuring a parameter of a roadway bolt, characterized by, include: Acquire three-dimensional point cloud data within the tunnel, and based on the three-dimensional point cloud data, determine the point cloud sub-data corresponding to each anchor bolt; For each anchor bolt, the point cloud sub-data is subjected to axis enhancement extraction to obtain the target axis reference point and the target axis unit direction vector corresponding to the anchor bolt. Based on the point cloud sub-data, the target axis reference point and the target axis unit direction vector, the exposed start point and exposed end point of the anchor bolt are determined; Based on the target axis reference point, the target axis unit direction vector, the exposed start point, and the exposed end point, the exposed parameters corresponding to the anchor bolt are determined.

2. The method of claim 1, wherein, The step of determining the point cloud sub-data corresponding to each anchor rod based on the three-dimensional point cloud data includes: The three-dimensional point cloud data is preprocessed, and the pre-trained point cloud semantic segmentation model is used to classify the preprocessed three-dimensional point cloud data to obtain anchor bolt type point cloud data. The anchor bolt point cloud data is clustered to obtain the point cloud sub-data corresponding to each anchor bolt.

3. The method of claim 1, wherein, The step of performing axis enhancement extraction on the point cloud sub-data to obtain the target axis reference point and the target axis unit direction vector corresponding to the anchor bolt includes: The point cloud sub-data is fitted with an axis to obtain an initial axis reference point and an initial axis unit direction vector; Axial analysis and extraction are performed on the point cloud sub-data along the unit direction vector of the initial axis to obtain the main trunk center point cloud sub-data corresponding to the anchor bolt; Data augmentation is performed on the cloud sub-data of the trunk center point, and axis fitting is performed on the cloud sub-data of the trunk center point after data augmentation to obtain the target axis reference point and the target axis unit direction vector.

4. The method of claim 1, wherein, The determination of the exposed start and end points of the anchor bolt based on the point cloud sub-data, the target axis reference point, and the target axis unit direction vector includes: The point cloud sub-data is projected along the unit direction vector of the target axis to obtain a one-dimensional projection sequence containing multiple projection values. Discretize the one-dimensional projection sequence to obtain an interval sequence containing multiple axial intervals; Based on the target axis reference point and the target axis unit direction vector, determine the corresponding weighted point density for each axial interval in the interval sequence; Based on the corresponding weighted point density of each axial interval, an exposure determination is performed on each axial interval in the interval sequence to obtain at least one candidate exposed interval. For each candidate exposed section, a local axis reference point corresponding to the candidate exposed section is determined; Based on the target axis unit direction vector and the local axis reference point, determine the average direction deviation angle of the candidate exposed interval relative to the target axis unit direction vector; Based on the average directional deviation angle corresponding to each candidate exposed interval, at least one exposed interval is selected from all candidate exposed intervals. Based on the target axis reference point, the target axis unit direction vector, the corresponding weighted point density of each axial interval, and the average direction deviation angle corresponding to each candidate exposed interval, the exposed starting point of the anchor rod is determined. The maximum projection value is found in all exposed sections, and the maximum projection value is determined as the exposed end point corresponding to the anchor rod.

5. The method of claim 4, wherein, The exposure determination is performed on each axial interval in the interval sequence based on the corresponding weighted point density of each axial interval to obtain at least one candidate exposed interval, including: Based on the weighted point density corresponding to each axial interval, the change in weighted point density corresponding to each pair of adjacent axial intervals in the interval sequence is determined. Determine the density change threshold based on the weighted point density corresponding to all axial intervals; For each pair of adjacent axial intervals, in response to determining that the weighted point density change is less than or equal to the density change threshold, each axial interval corresponding to the pair of adjacent axial intervals is selected as a candidate exposed interval.

6. The method of claim 4, wherein, The step of selecting at least one exposed interval from all candidate exposed intervals based on the average directional deviation angle corresponding to each candidate exposed interval includes: For each candidate exposed interval, in response to determining that the average directional deviation angle corresponding to the candidate exposed interval is less than or equal to the directional consistency threshold, the candidate exposed interval is determined as the exposed interval.

7. The method of claim 4, wherein, The process of determining the exposed starting point of the anchor bolt against the wall based on the target axis reference point, the unit direction vector of the target axis, the weighted point density corresponding to each axial interval, and the average direction deviation angle corresponding to each candidate exposed interval includes: For each axial interval, the weighted average off-axis scale corresponding to the axial interval is determined based on the target axis reference point and the target axis unit direction vector; In the weighted average off-axis scale corresponding to all axial intervals, find the weighted average off-axis scale corresponding to each exposed interval, and calculate the mean of the weighted average off-axis scale corresponding to all exposed intervals. Based on the mean, a stable rod-shaped reference scale is determined; In the weighted point density corresponding to each axial interval, find the weighted point density corresponding to each exposed interval; and in the average direction deviation angle corresponding to each candidate exposed interval, find the average direction deviation angle corresponding to each exposed interval. For each exposed interval, the stability score corresponding to the exposed interval is determined based on the weighted point density corresponding to the exposed interval, the average directional deviation angle corresponding to the exposed interval, and the stable rod-shaped reference scale. In response to determining that the stability score is greater than or equal to the interval score threshold, the exposed interval is taken as the target axial interval; The minimum projection value is found in the entire target axial interval, and the minimum projection value is determined as the exposed starting point.

8. The method according to claim 7, characterized in that, The stability score corresponding to the exposed interval is determined based on the weighted point density corresponding to the exposed interval, the average directional deviation angle corresponding to the exposed interval, and the stable rod-shaped reference scale, including: The weighted point density corresponding to the exposed interval is normalized to obtain the weighted point density normalized value; The average direction deviation angle score is determined based on the average direction deviation angle corresponding to the exposed area. Based on the weighted average off-axis scale and stable rod-shaped reference scale corresponding to the exposed interval, the off-axis scale consistency score is determined. The stability score is determined based on the weighted point density normalized value, the average direction deviation angle score, and the off-axis scale consistency score.

9. The method according to claim 1, characterized in that, The determination of the exposed parameters corresponding to the anchor bolt based on the target axis reference point, the unit direction vector of the target axis, the exposed start point, and the exposed end point includes: Obtain the axis data of the tunnel, and determine the point with the smallest distance from the exposed starting point in the axis data as the closest point; Obtain other data within a predetermined range from the nearest point from the axis data; Based on the nearest point and the other data, the direction of the local tunnel axis is determined; Based on the direction of the local tunnel axis, determine the tunnel cross-section reference plane; The target axis unit direction vector is projected onto the tunnel cross-section reference plane to obtain the in-plane projection vector, and the in-plane projection vector is normalized to obtain the unit radial vector. Based on the angle between the unit direction vector of the target axis and the unit radial vector, the radial deflection angle corresponding to the anchor rod is determined; The exposed length of the anchor rod is determined based on the exposed starting point and the exposed ending point. The radial deflection angle and the exposed length of the anchor rod are used as the exposed parameters of the anchor rod.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 9.

Images