Tunnel scene point cloud scanning scheme optimization method considering area priority evaluation

Abstract

A tunnel scene point cloud scanning scheme optimization method considering area priority evaluation comprises the following steps: 1) constructing a three-dimensional solution space model of a target tunnel based on a two-dimensional design drawing and a three-dimensional BIM design model of a target tunnel scene; 2) performing grid division on the three-dimensional solution space model of the target tunnel by adopting a grid discrete function of BIM software, and performing center point coordinate calculation based on divided grids to obtain a tunnel scanning object set; 3) performing priority division on the sections of the target tunnel according to corresponding construction safety risk assessment guidelines and specifications; 4) constructing a scanning site hierarchical scanning model based on the tunnel scanning object set and the divided section priority; according to the long-line tunnel point cloud optimization acquisition method, a technical reference is provided for data redundancy suppression and comprehensive optimization of long-line tunnel point cloud optimization acquisition, and the long-line tunnel point cloud optimization acquisition method comprises the following steps of (1) obtaining a scanning site grading scanning model, (2) solving the scanning site grading scanning model by utilizing a genetic algorithm to obtain an optimal scanning scheme, and (3) arranging a scanner in a target tunnel based on the optimal scanning scheme to acquire point cloud data.

Description

Technical Field

[0001] This invention relates to the fields of information technology and building safety management technology, specifically to an optimization method for point cloud scanning schemes in tunnel scenes that considers regional priority assessment. Background Technology

[0002] Tunnels are long, linear structures built underground or passing through mountains, typically used for transportation, water conveyance, or mining. During tunnel operation and maintenance, the support structure is at risk of deformation and damage due to the time-varying creep characteristics of the surrounding rock and other factors. To ensure tunnel structural safety, regular monitoring of its full-section spatial deformation is essential. However, current measurement methods using equipment such as total stations and inclinometers are mainly used to calculate deformation at specific monitoring points, which is insufficient for a comprehensive and accurate assessment of the overall tunnel structure. Therefore, in recent years, researchers have proposed a solution: using scanning equipment such as terrestrial laser scanners to acquire high-precision point cloud data of the tunnel, and then generating a complete and accurate tunnel model through registration and modeling. Simultaneously, to minimize the number of scans and generate a more accurate and complete point cloud, it is necessary to optimize the scanning scheme. However, current scanning scheme optimization methods still have certain limitations: on the one hand, they fail to fully consider the large scanning workload and massive amount of scanned data caused by the long linear structure of tunnels during data acquisition; on the other hand, most optimization methods only consider scanning time and coverage for single optimization solutions, failing to take scanning resolution as a constraint for further comprehensive optimization. These shortcomings in considering actual engineering conditions mean that point cloud scanning scheme optimization has not yet achieved optimal results and cannot fully meet the needs of engineering practice.

[0003] Therefore, in order to address the core issues of current point cloud scanning scheme optimization methods that ignore the redundancy of scanning data and fail to conduct comprehensive optimization from the perspective of long linear scanning space, it is urgent to establish a point cloud scanning scheme optimization method for tunnel scenarios that considers regional priority assessment, so as to improve the effectiveness and engineering applicability of scanning scheme optimization. Summary of the Invention

[0004] The purpose of this invention is to provide an optimization method for point cloud scanning schemes in tunnel scenes that considers regional priority assessment, including:

[0005] 1) Based on the two-dimensional design drawings and three-dimensional BIM design model of the target tunnel scene, construct a three-dimensional solution space model of the target tunnel.

[0006] 2) The BIM software's mesh discretization function is used to divide the three-dimensional solution space model of the target tunnel into a mesh, and the center point coordinates are calculated based on the divided mesh to obtain the tunnel scanning object set.

[0007] 3) Prioritize the sections of the target tunnel by referring to the relevant construction safety risk assessment guidelines and specifications.

[0008] 4) Based on the set of tunnel scanning objects and the priority of the divided sections, a hierarchical scanning model for scanning sites is constructed.

[0009] 5) Use a genetic algorithm to solve the hierarchical scanning model of the scanning sites to obtain the optimal scanning scheme.

[0010] 6) Based on the optimal scanning scheme, the scanner is placed inside the target tunnel to collect point cloud data.

[0011] Furthermore, the three-dimensional solution space model of the target tunnel includes the secondary lining inner wall, the road surface, and the drainage ditch cover.

[0012] Furthermore, the steps for obtaining the tunnel scanning object set are as follows:

[0013] A1 uses the mesh discretization function of BIM software to divide the inner wall of the secondary lining into a mesh, and obtains the coordinates of the four vertices of the divided mesh.

[0014] A2 calculates the coordinates of the center point of the grid based on the coordinates of the four vertices of the divided grid, and uses this as the target point for scanning.

[0015] The coordinates of the center point of the grid are shown below:

[0016] (1)

[0017] In the formula, Indicates grid index, Indicates the first The coordinates of the center point of each grid. , , , The first The coordinates of the four vertices of a grid.

[0018] A3 selects coordinate points at preset intervals on the center axis of the road surface as candidate scanning sites.

[0019] A4 constructs a set of tunnel scanning objects based on the target scanning points and candidate scanning sites.

[0020] Furthermore, the prioritization of the target tunnel sections is based on the risk scores for tunnel collapse accidents and water inrush and mudslide accidents, as shown below:

[0021] (2)

[0022] (3)

[0023] (4)

[0024] In the formula, This indicates the risk score for a tunnel collapse accident. This indicates the risk score for water inrush and mudslide accidents. This represents the reduction factor. This indicates the score of the surrounding rock grade index. The score indicates the fault rupture condition. The score represents the indicative value of the seepage status. This indicates the score for geological compliance index. This indicates the score for the construction method indicator. This indicates the score for the construction step distance index. The score represents the degree of karst development. This indicates the score for the fault rupture zone index. The score indicates the condition of the surrounding water body. This indicates the score for each evaluation indicator.

[0025] Furthermore, the objective function of the hierarchical scanning model for scanning sites is as follows:

[0026] (5)

[0027] In the formula, This represents the objective function of the hierarchical scanning model for scanning sites. This indicates the index of candidate scan sites. Indicates from the first A set of target points captured by candidate scanning sites. This represents the total number of target points captured from each candidate scan site. This indicates the total number of target points scanned.

[0028] Furthermore, the constraints of the hierarchical scanning model of the scanning site include scanning distance constraints, occlusion constraints, quality constraints of the scanned object, level of detail constraints, and overlapping area constraints.

[0029] Furthermore, the scanning distance constraint is as follows:

[0030] (6)

[0031] (7)

[0032] In the formula, This indicates the scanning distance from the candidate scanning site to the target scanning point. This indicates the scanning radius of the scanner. This indicates the coordinates of the target point being scanned. Indicates the coordinates of the candidate scan site.

[0033] The occlusion constraint adds a candidate scanning station to the area where scanning is obstructed within the tunnel.

[0034] The area where the scan was blocked included the emergency parking lane.

[0035] The quality constraints of the scanned object are as follows:

[0036] (8)

[0037] (9)

[0038] In the formula, This indicates the angle of incidence of the target point being scanned. This represents the line-of-sight vector from the target point to the candidate scanning site. This represents the normal vector of the target point being scanned. Represents the norm. Indicates the limit scanning angle.

[0039] The level-of-detail constraints are as follows:

[0040] (10)

[0041] (11)

[0042] (12)

[0043] In the formula, This indicates the horizontal distance between two adjacent scan target points. Indicates the horizontal angle of incidence. This indicates the angular resolution; different priority segments have different angular resolutions. This indicates the distance between the candidate scanning site and the target scanning point. This represents the vertical distance between two adjacent scan target points. This indicates the angle of incidence. Indicates the first A set of target points captured by candidate scanning sites. This indicates the level of detail required for horizontal scanning. This indicates the level of detail required for vertical scanning. This indicates the coordinates of the target point being scanned.

[0044] The area constraint of the overlapping region is as follows:

[0045] (13)

[0046] (14)

[0047] (15)

[0048] In the formula, This indicates the scanning radius of the scanner. This indicates the width of the road surface. It represents the complementary angle of the incident angle at the farthest scanned target point. This represents the distance between two candidate scan sites. Indicates the length of the overlapping region. This indicates the area of ​​the overlapping region.

[0049] Furthermore, when using the genetic algorithm to solve the hierarchical scanning model for scanning sites, a scanning time constraint is also introduced. The hierarchical scanning model for scanning sites is then shown below:

[0050] (16)

[0051] (17)

[0052] In the formula, This represents the maximization function. This represents the objective function of the hierarchical scanning model for scanning sites. Indicates the scan time. This represents the maximum constraint value for scan time. This indicates the index of candidate scan sites. This indicates the total number of candidate scanning sites. Indicates the first The candidate scanning sites and the first Overlapping area between candidate scan sites. This indicates the time it takes to move the scanner to the next candidate scanning site. This indicates the scan resolution index. This indicates the total number of scan resolutions. Indicates the scanner at the The candidate scanning site with the first The time consumed when performing a scan at each scan resolution.

[0053] Furthermore, the scanner includes a laser scanner.

[0054] Furthermore, the quality of the acquired point cloud data is evaluated by local point density using the point cloud data acquired by a scanner deployed according to the optimal scanning scheme, as shown below:

[0055] (18)

[0056] In the formula, This represents the local point density. This represents the number of points in the neighborhood of the point cloud data. This represents the neighborhood radius of the point cloud data to the nearest neighbor.

[0057] The technical effect of this invention is undeniable. The tunnel scene point cloud scanning scheme optimization method provided by this invention, which considers regional priority evaluation, can take into account the priority level of different regions of the tunnel, and meet the scanning data quality requirements of different regions by setting different scanning resolutions. It establishes a solution space model of the target scene to perform scanning constraint calculations on the tunnel, and finally achieves scanning scheme optimization with scanning coverage rate as the objective function.

[0058] This invention addresses the problems in current point cloud data scanning scheme optimization techniques that ignore scanning data redundancy and fail to optimize from the perspective of long linear scanning space, providing a technical reference for suppressing and comprehensively optimizing data redundancy in long linear tunnel point cloud optimization acquisition. Attached Figure Description

[0059] Figure 1 A flowchart of an optimization method for point cloud scanning schemes in tunnel scenes that considers regional priority assessment;

[0060] Figure 2 This is a schematic diagram of tunnel model preprocessing; Figure 2 (a) is a schematic diagram of a 3D BIM model; Figure 2 (b) is a schematic diagram of secondary lining; Figure 2 (c) is a schematic diagram of the road surface; Figure 2 (d) is a schematic diagram of the drainage ditch cover; Figure 2 (e) is a schematic diagram of the solution space model;

[0061] Figure 3 This is a schematic diagram of the mesh generation results for the tunnel model;

[0062] Figure 4 This is a schematic diagram of the target points and candidate scanning locations for tunnel scanning.

[0063] Figure 5 A schematic diagram illustrating the priority assessment of different areas within the tunnel;

[0064] Figure 6 This is a schematic diagram illustrating the scanning distance filtering process.

[0065] Figure 7 A schematic diagram of the tunnel cross-section variation and obstruction analysis; Figure 7 (a) is a schematic diagram of occlusion analysis from a three-dimensional perspective; Figure 7 (b) is a schematic diagram of planar viewpoint occlusion analysis;

[0066] Figure 8 This is a schematic diagram illustrating the calculation of the incident angle during scanning in a tunnel scene.

[0067] Figure 9 This is a schematic diagram illustrating the geometric relationship between scan parameters and level of detail.

[0068] Figure 10 This is a schematic diagram showing the distribution of target spheres between adjacent scanning stations;

[0069] Figure 11 A schematic diagram illustrating the division of tunnel segments into different levels;

[0070] Figure 12 This is the convergence graph for the genetic algorithm optimization.

[0071] Figure 13 This is a schematic diagram of a scanning optimization scheme for a tunnel scene.

[0072] Figure 14 This is a schematic diagram of the point density analysis results for the tunnel point cloud. Detailed Implementation

[0073] The present invention will be further described below with reference to embodiments, but it should not be construed that the scope of the present invention is limited to the following embodiments. Various substitutions and modifications made based on ordinary technical knowledge and common practices in the art without departing from the above-described technical concept of the present invention should be included within the scope of protection of the present invention.

[0074] Example 1:

[0075] See Figures 1 to 14 An optimization method for point cloud scanning schemes in tunnel scenes considering regional priority assessment includes:

[0076] 1) Based on the two-dimensional design drawings and three-dimensional BIM design model of the target tunnel scene, construct a three-dimensional solution space model of the target tunnel.

[0077] 2) The BIM software's mesh discretization function is used to divide the three-dimensional solution space model of the target tunnel into a mesh, and the center point coordinates are calculated based on the divided mesh to obtain the tunnel scanning object set.

[0078] 3) Prioritize the sections of the target tunnel by referring to the relevant construction safety risk assessment guidelines and specifications.

[0079] 4) Based on the set of tunnel scanning objects and the priority of the divided sections, a hierarchical scanning model for scanning sites is constructed.

[0080] 5) Use a genetic algorithm to solve the hierarchical scanning model of the scanning sites to obtain the optimal scanning scheme.

[0081] 6) Based on the optimal scanning scheme, the scanner is placed inside the target tunnel to collect point cloud data.

[0082] Example 2:

[0083] An optimization method for point cloud scanning scheme of tunnel scene considering regional priority assessment is described in Example 1. Further, the three-dimensional solution space model of the target tunnel includes the secondary lining inner wall, road surface, and drainage ditch cover.

[0084] Example 3:

[0085] An optimization method for point cloud scanning scheme in tunnel scenes considering regional priority assessment, the main technical contents of which are described in any one of Embodiments 1 and 2, further wherein the step of obtaining the tunnel scanning object set is as follows:

[0086] A1 uses the mesh discretization function of BIM software to divide the inner wall of the secondary lining into a mesh, and obtains the coordinates of the four vertices of the divided mesh.

[0087] A2 calculates the coordinates of the center point of the grid based on the coordinates of the four vertices of the divided grid, and uses this as the target point for scanning.

[0088] The coordinates of the center point of the grid are shown below:

[0089] (1)

[0090] In the formula, Indicates grid index, Indicates the first The coordinates of the center point of each grid. , , , The first The coordinates of the four vertices of a grid.

[0091] A3 selects coordinate points at preset intervals on the center axis of the road surface as candidate scanning sites.

[0092] A4 constructs a set of tunnel scanning objects based on the target scanning points and candidate scanning sites.

[0093] Example 4:

[0094] An optimization method for point cloud scanning schemes in tunnel scenes considering regional priority assessment is provided. The main technical content is described in any one of Examples 1 to 3. Further, the priority division of the target tunnel sections is based on the risk scores of tunnel collapse accidents and water inrush and mudslide accidents, as shown below:

[0095] (2)

[0096] (3)

[0097] (4)

[0098] In the formula, This indicates the risk score for a tunnel collapse accident. This indicates the risk score for water inrush and mudslide accidents. This represents the reduction factor. This indicates the score of the surrounding rock grade index. The score indicates the fault rupture condition. The score represents the indicative value of the seepage status. This indicates the score for geological compliance index. This indicates the score for the construction method indicator. This indicates the score for the construction step distance index. The score represents the degree of karst development. This indicates the score for the fault rupture zone index. The score indicates the condition of the surrounding water body. This indicates the score for each evaluation indicator.

[0099] Based on the calculated risk scores and the risk event probability level assessment table in the guidelines, the tunnels are divided into three levels: I, II, and III.

[0100] For example, a tunnel section classified according to its surrounding rock grade is classified as Grade IV, with a small-scale fault fracture zone and relatively developed karst with karst fissure zones. The calculated risk score for the tunnel section's collapse accident is 7 points, corresponding to a risk level of Grade III. The risk score for the tunnel's water inrush and mudslide accident is 2 points, corresponding to a risk level of Grade II. Therefore, the tunnel section is prioritized according to the higher risk level of Grade III.

[0101] The higher the risk score, the higher the risk level, and higher-risk areas require higher priority and more precise scanning.

[0102] Example 5:

[0103] An optimization method for point cloud scanning schemes in tunnel scenes considering regional priority assessment is described in any one of Examples 1 to 4. Further, the objective function of the hierarchical scanning model for the scanning sites is as follows:

[0104] (5)

[0105] In the formula, This represents the objective function of the hierarchical scanning model for scanning sites. This indicates the index of candidate scan sites. Indicates from the first A set of target points captured by candidate scanning sites. This represents the total number of target points captured from each candidate scan site. This indicates the total number of target points scanned.

[0106] Example 6:

[0107] An optimization method for point cloud scanning scheme in tunnel scenes considering regional priority assessment is provided. The main technical contents are described in any one of Examples 1 to 5. Furthermore, the constraints of the hierarchical scanning model of the scanning site include scanning distance constraints, occlusion constraints, quality constraints of the scanned object, level of detail constraints, and area constraints of the overlapping region.

[0108] Example 7:

[0109] An optimization method for point cloud scanning scheme in tunnel scenes considering regional priority assessment, the main technical contents of which are described in any one of Embodiments 1 to 6, and further, the scanning distance constraint is as follows:

[0110] (6)

[0111] (7)

[0112] In the formula, This indicates the scanning distance from the candidate scanning site to the target scanning point. This indicates the scanning radius of the scanner. This indicates the coordinates of the target point being scanned. Indicates the coordinates of the candidate scan site.

[0113] The occlusion constraint adds a candidate scanning station to the area where scanning is obstructed within the tunnel.

[0114] The area where the scan was blocked included the emergency parking lane.

[0115] The quality constraints of the scanned object are as follows:

[0116] (8)

[0117] (9)

[0118] In the formula, This indicates the angle of incidence of the target point being scanned. This represents the line-of-sight vector from the target point to the candidate scanning site. This represents the normal vector of the target point being scanned. Represents the norm. Indicates the limit scanning angle.

[0119] The level-of-detail constraints are as follows:

[0120] (10)

[0121] (11)

[0122] (12)

[0123] In the formula, This indicates the horizontal distance between two adjacent scan target points. Indicates the horizontal angle of incidence. This indicates the angular resolution; different priority segments have different angular resolutions. This indicates the distance between the candidate scanning site and the target scanning point. This represents the vertical distance between two adjacent scan target points. This indicates the angle of incidence. Indicates the first A set of target points captured by candidate scanning sites. This indicates the level of detail required for horizontal scanning. This indicates the level of detail required for vertical scanning. This indicates the coordinates of the target point being scanned.

[0124] Different priorities correspond to different accuracy requirements, thereby enabling targeted scanning through different angular resolutions.

[0125] The area constraint of the overlapping region is as follows:

[0126] (13)

[0127] (14)

[0128] (15)

[0129] In the formula, This indicates the scanning radius of the scanner. This indicates the width of the road surface. It represents the complementary angle of the incident angle at the farthest scanned target point. This represents the distance between two candidate scan sites. Indicates the length of the overlapping region. This indicates the area of ​​the overlapping region.

[0130] Example 8:

[0131] An optimization method for point cloud scanning scheme in tunnel scenes considering regional priority assessment is provided. The main technical content is described in any one of Examples 1 to 7. Furthermore, when using a genetic algorithm to solve the hierarchical scanning model for scanning sites, a scanning time constraint is introduced. The hierarchical scanning model for scanning sites is then shown below:

[0132] (16)

[0133] (17)

[0134] In the formula, This represents the maximization function. This represents the objective function of the hierarchical scanning model for scanning sites. Indicates the scan time. This represents the maximum constraint value for scan time. This indicates the index of candidate scan sites. This indicates the total number of candidate scanning sites. Indicates the first The candidate scanning sites and the first Overlapping area between candidate scan sites. This indicates the time it takes to move the scanner to the next candidate scanning site. This indicates the scan resolution index. This indicates the total number of scan resolutions. Indicates the scanner at the The candidate scanning site with the first The time consumed when performing a scan at each scan resolution.

[0135] The genetic algorithm first generates chromosomes using a hybrid encoding method, taking into account both scan position and scan resolution. Then, it calculates the coverage of the current scheme by combining scan constraints. Finally, it iteratively solves the optimal scan scheme that meets the coverage requirements through selection, crossover, and mutation operations.

[0136] Example 9:

[0137] An optimization method for point cloud scanning scheme in tunnel scenes that considers regional priority assessment is provided. The main technical contents are described in any one of Embodiments 1 to 8. Further, the scanner includes a laser scanner.

[0138] Example 10:

[0139] A method for optimizing a point cloud scanning scheme in a tunnel scene, considering regional priority assessment, is described in any one of Examples 1 to 9. Further, the quality of the acquired data is evaluated by local point density using point cloud data collected by scanners arranged according to the optimal scanning scheme, as shown below:

[0140] (18)

[0141] In the formula, This represents the local point density. This represents the number of points in the neighborhood of the point cloud data. This represents the neighborhood radius of the point cloud data to the nearest neighbor.

[0142] Example 11:

[0143] See Figures 1 to 14 An optimization method for point cloud scanning schemes in tunnel scenes considering regional priority assessment, the main technical contents of which include:

[0144] S101. Based on the two-dimensional design drawings and three-dimensional BIM design model of the target tunnel scenario, construct a three-dimensional solution space model of the tunnel.

[0145] S102. For the tunnel inner contour model obtained in step S101, firstly, the secondary lining is divided into grids using the grid discretization function of BIM software, and then the center point coordinates are calculated based on the obtained grid to obtain the set of scanning target points.

[0146] S103. For the tunnel inner contour model obtained in step S101, the road surface is first divided into grids using the grid discretization function of BIM software. Then, the center point coordinates are calculated based on the obtained grids to obtain a set of candidate scanning positions. This set of candidate scanning positions is then used to extract data from the set of scanning target points obtained in step S102.

[0147] S104. Based on the engineering geological and hydrogeological conditions report and structural status of the target tunnel, prioritize and classify the tunnel sections in accordance with the relevant construction safety risk assessment guidelines and specifications.

[0148] S105. Based on the tunnel scanning object set extracted in steps S102 and S103, a visibility analysis is performed on each scanning station. First, the scanning distance of each scanning station is filtered, and then the cross-sectional change occlusion analysis of the tunnel scene is performed.

[0149] S106. Based on the tunnel scanning object set extracted in steps S102 and S103, perform an accuracy analysis on each scanning target point that each scanning station can capture, and constrain the quality of the scanning object by controlling the incident angle and scanning distance.

[0150] S107. Based on the tunnel scanning object set extracted in steps S102 and S103, and the priority evaluation and division of different tunnel regions in S104, the scanning quality requirements of each graded region are first defined, and then the detailed level constraints of each scanning target point are analyzed.

[0151] S108. Based on the set of tunnel scanning objects extracted in steps S102 and S103, perform a registration analysis between two adjacent scans.

[0152] S109. Based on the priority evaluation and division method in step S104, a genetic algorithm is used to optimize the optimal scanning scheme for point cloud data acquisition of the target tunnel, so as to capture the most complete scanned object with the fewest possible scanning stations while meeting the data quality requirements.

[0153] S110. Based on the target tunnel scanning optimization scheme obtained in step S109, point cloud data is collected using a land laser scanner, and then the point density analysis of the scan data is performed for verification.

[0154] Step S101 includes:

[0155] First, the tunnel model needs to be preliminarily processed. Combining the 2D drawings and the layered structure of the BIM model, the various components of the BIM model are disassembled. Then, the external structures such as the initial support and redundant auxiliary components are manually deleted, leaving only the secondary lining inner wall, road surface, drainage ditch and cable trench cover surface, etc., to form the scanned inner contour model and determine the workspace for scan optimization.

[0156] Step S102 includes:

[0157] First, based on the tunnel's inner contour BIM model, the inner wall surface of the secondary lining is divided into multiple rectangular meshes using a mesh discretization function. The rectangular meshes of the secondary lining are connected by four vertices. , , , The three-dimensional coordinates of the grid are used to calculate the coordinates of the center point of the rectangular grid, which serves as the target point for subsequent evaluation of the grid's data quality. (Center point coordinates of any rectangular grid) It can be represented as

[0158]

[0159] Step S103 includes:

[0160] First, based on the tunnel's inner contour BIM model, the inner wall surface of the secondary lining is divided into multiple rectangular grids using a mesh discretization function. Then, the road surface grid is divided at 10m intervals to obtain candidate scan location grids with 10m intervals. Next, the center point of each segment of the road surface is taken, and the vertex coordinates of the rectangular grid at the center of the road surface are used. , The calculated or directly represented interval points form a candidate scan site set, which together with the scan target point set obtained in step S102 constitutes the scan object set.

[0161] Step S104 includes:

[0162] The guidelines use section risk assessment of tunnels as an evaluation indicator for importance zoning, and then employ an indicator system method to estimate the probability of accidents. Based on the calculated risk scores and the risk event probability level assessment table in the guidelines, tunnels are divided into three levels: I, II, and III. The risk score for tunnel collapse accidents is... Risk score for water inrush and mud inrush accidents It can be calculated as follows:

[0163]

[0164]

[0165]

[0166] in, This is a reduction factor that takes into account the possibility of accidents caused by human factors and construction management. It consists of the scores for each evaluation indicator, and then... It is the score of the surrounding rock grade index. It is the score of the fault rupture condition index. It is the score of the seepage status index. It is the geological compliance index score. It is the score for construction method indicators. It is the score for the construction step distance index. It is a score indicating the degree of karst development. It is the index score of the fault fracture zone. It is the score of the surrounding water condition index.

[0167] Step S105 includes:

[0168] First, the scanning stations need to be screened based on scanning distance. This means the straight-line distance between the object to be scanned and the candidate scanning station should be less than the set maximum scanning distance. The scanning distance from the target point to the scanner... It should be smaller than the radius of the scan range. .

[0169]

[0170] in, These are the coordinates of the target point. These are the scanner coordinates; It is in the The set of target points that meet the criteria captured by each candidate scan location. The occlusion analysis for tunnels is due to the fact that emergency stopping zones within tunnels may obstruct the scan. In this case, to avoid discontinuities in the overall scan data, a separate scan should be performed in the emergency stopping zone section.

[0171] Step S106 includes:

[0172] First, the maximum scanning distance and scanning angle are set based on the actual size of the target tunnel. As the scanning distance and incident angle increase, the measurement accuracy decreases. Since the scanning distance is constrained by visibility analysis, the scanning angle also needs to be limited. Then, the normal vector of the target point is calculated. With line of sight vector The angle between them can be used to obtain the angle of incidence of the target point in the scene. :

[0173]

[0174] Step S107 includes:

[0175] First, based on the risk assessment results, the required scanning precision for different areas is determined according to the intended use of the scanned data. Different areas have different requirements for scan detail level, necessitating the use of different angular resolutions to meet the data quality requirements of each area. To simplify the calculation process, the distance between scan points is decomposed into the distance between two adjacent scan points along the horizontal and vertical directions. By comparing this distance with the relevant scan detail level requirements, target points meeting the point density requirements can be selected. Therefore, the first... A set of target points that meet the requirements at candidate scan locations. :

[0176]

[0177]

[0178]

[0179] in, It is the horizontal distance between two adjacent target points; It is the vertical distance; It is angular resolution. It is the distance between the scan location and the target point. and These represent the horizontal and vertical angles of incidence, respectively. It is a requirement for horizontal scanning detail level. It is a requirement for vertical scanning of detailed levels.

[0180] Step S108 includes:

[0181] First, the scanning range for each station is determined based on the limit scanning distance and target tunnel size information set in step S105. To ensure data registration accuracy, there should be a certain overlap between two adjacent scanning positions to ensure that there are at least three common artificial targets as the basis for alignment between two scans. The area of ​​the overlap region... The calculation is as follows:

[0182]

[0183]

[0184]

[0185] in, It is the scanner's scanning distance. It is the complementary angle of the angle of incidence at the farthest target. It is the width of the tunnel surface. It is the distance between two scan positions. It is the length of the overlapping region.

[0186] Step S109 includes:

[0187] First, the target tunnel is divided into regional hierarchies, and then the objective function is set. This ensures that the tunnel point cloud is as complete as possible in different priority partitions.

[0188]

[0189] in, Indicates from the first Target points captured at each scanning location; This represents the total number of visible target points obtained from each selected scan location; It represents the total number of target points scanned.

[0190] Next, to prevent the optimization results from getting trapped in local optima, in addition to the aforementioned constraint on scan coverage, a constraint is imposed on the scan time. The maximum constraint value for the scan time is set... Let the scan time be set to the greedy algorithm's scan time within the same scene. Therefore, the optimization problem can be described as follows:

[0191]

[0192]

[0193] in, Indicates the first The second scan and the first +1 overlapping area between scans; This indicates the number of scan locations in the scan plan; It is the number of available scanning angular resolutions; This indicates the time required to move the scanner to the next location; Indicates the scanner at the The scan position is the first The time consumed during scanning at each scan resolution. Subsequently, during optimization, chromosomes are generated using a combination of binary and integer encoding methods, and the SEGA algorithm is used for optimization.

[0194] Step S110 includes:

[0195] First, based on the target tunnel scanning optimization scheme obtained in step 109, point cloud data of the tunnel is collected according to the scanning position and scanning resolution in the scheme. Then, the point clouds acquired from each station are registered one by one and integrated to generate the overall point cloud data of the target tunnel. Next, the local point density of the point cloud data is calculated to evaluate the data quality, using the following formula:

[0196]

[0197] in It is the number of points in the neighborhood of a point. It is the radius of the neighborhood of the nearest neighbor at that point.

[0198] Example 12:

[0199] See Figures 1 to 14 An optimization method for point cloud scanning schemes in tunnel scenes considering regional priority assessment, the main technical contents of which include:

[0200] S101. Based on the two-dimensional design drawings and three-dimensional BIM design model of the target tunnel scene, construct a three-dimensional solution space model of the tunnel.

[0201] In practice, the tunnel model needs to be preliminarily processed. Combining the 2D drawings and the layered structure of the BIM model, the various components of the BIM model are decomposed, such as... Figure 2 As shown in (a) to (d). Then, the initial support and other external structures and redundant auxiliary components are manually deleted, leaving only the inner wall of the secondary lining, the road surface, the drainage ditch, and the surface of the cable trench cover, etc., to form a scanned inner contour model, as shown below. Figure 2 As shown in (e), the workspace for scan optimization is determined.

[0202] S102. For the tunnel inner contour model obtained in step S101, the secondary lining is first divided into grids using the grid discretization function of BIM software. Then, the center point coordinates are calculated based on the obtained grid to obtain the set of scanning target points.

[0203] In practice, based on the tunnel's inner contour BIM model, the surface of the secondary lining's inner wall is first divided into multiple rectangular grids using a mesh discretization function. The rectangular grids of the secondary lining are connected by four vertices. , , , The three-dimensional coordinates of the grid are used to calculate the coordinates of the center point of the rectangular grid, which serves as the target scanning point for subsequent evaluation of the grid's data quality. Figure 3 As shown. Coordinates of the center point of an arbitrary rectangular grid. It can be represented as

[0204] S103. For the tunnel inner contour model obtained in step S101, the road surface is first divided into grids using the grid discretization function of BIM software. Then, the center point coordinates are calculated based on the obtained grids to obtain a set of candidate scanning positions. This set of candidate scanning positions is then used to extract data from the set of scanning target points obtained in step S102.

[0205] In practice, firstly, based on the tunnel's inner contour BIM model, the inner wall surface of the secondary lining is divided into multiple rectangular grids using a mesh discretization function. Then, the road surface grid is divided at 10m intervals to obtain candidate scan position grids with 10m intervals. Next, the center point of each segment of the road surface is taken, and the vertex coordinates of the rectangular grid at the center of the road surface are used. , The calculated or directly represented interval points form a candidate scan site set, such as Figure 3 As shown, together with the set of scan target points obtained in step S102, they constitute the scan object set, as follows: Figure 4 As shown.

[0206] S104. Based on the engineering geological and hydrogeological conditions report and structural status of the target tunnel, prioritize and classify the tunnel sections in accordance with the relevant construction safety risk assessment guidelines and specifications.

[0207] In practice, the guidelines use the risk assessment of tunnel sections as the basis for evaluating the importance of zoning, and then employ an indicator system method to estimate the probability of accidents. Based on the calculated risk scores and the risk event probability level assessment table in the guidelines, tunnels are divided into three levels: I, II, and III. Figure 5 As shown. The risk score for tunnel collapse accidents is... Risk score for water inrush and mud inrush accidents It can be calculated as follows:

[0208]

[0209]

[0210]

[0211] in, This is a reduction factor that takes into account the possibility of accidents caused by human factors and construction management. It consists of the scores for each evaluation indicator, and then... It is the score of the surrounding rock grade index. It is the score of the fault rupture condition index. It is the score of the seepage status index. It is the geological compliance index score. It is the score for construction method indicators. It is the score for the construction step distance index. It is a score indicating the degree of karst development. It is the index score of the fault fracture zone. It is the score of the surrounding water condition index.

[0212] S105. Based on the tunnel scanning object set extracted in steps S102 and S103, a visibility analysis is performed on each scanning station. First, the scanning distance of each scanning station is filtered, and then the cross-sectional change occlusion analysis of the tunnel scene is performed.

[0213] In practice, the first step is to filter the scanning stations based on their scanning distance. Specifically, the straight-line distance between the object to be scanned and the candidate scanning station should be less than the set maximum scanning distance. Figure 6 As shown, the scanning distance from the target point to the scanner It should be smaller than the radius of the scan range. .

[0214]

[0215] in, These are the coordinates of the target point. These are the scanner coordinates; It is in the The set of target points that meet the criteria captured at each candidate scan location. The tunnel occlusion analysis is due to the fact that emergency stopping lanes within the tunnel may obstruct the scan, such as... Figure 7 As shown. In this case, to avoid discontinuities in the overall scan data, a separate scan should be performed in the emergency stop zone.

[0216] S106. Based on the tunnel scanning object set extracted in steps S102 and S103, perform an accuracy analysis on each scanning target point that each scanning station can capture, and constrain the quality of the scanning object by controlling the incident angle and scanning distance.

[0217] In practice, the limiting scanning distance and scanning angle are first set according to the actual size of the target tunnel. As the scanning distance and incident angle increase, the measurement accuracy will continuously decrease. The scanning distance is constrained by the visibility analysis, and the scanning angle also needs to be limited. For example... Figure 8 As shown, the normal vector of the target point is calculated. With line of sight vector The angle between them can be used to obtain the angle of incidence of the target point in the scene. :

[0218]

[0219] S107. Based on the tunnel scanning object set extracted in steps S102 and S103, and the priority evaluation and division of different tunnel regions in S104, the scanning quality requirements of each graded region are first defined, and then the detailed level constraints of each scanning target point are analyzed.

[0220] In practice, the first step is to determine the required scanning precision for different areas based on the risk assessment results and the intended use of the scanned data. Different areas have varying requirements for scan detail levels, necessitating the use of different angular resolutions to meet the data quality requirements of each area. To simplify the calculation process, the distance between scan points is decomposed into the distance between two adjacent scan points along the horizontal and vertical directions. By comparing this distance with the relevant scan detail level requirements, target points that meet the point density requirements can be selected. For example... Figure 9 As shown, the first can be obtained A set of target points that meet the requirements at candidate scan locations. :

[0221]

[0222]

[0223]

[0224] in, It is the horizontal distance between two adjacent target points; It is the vertical distance; It is angular resolution. It is the distance between the scan location and the target point. and These represent the horizontal and vertical angles of incidence, respectively. It is a requirement for horizontal scanning detail level. It is a requirement for vertical scanning of detailed levels.

[0225] S108. Based on the set of tunnel scanning objects extracted in steps S102 and S103, perform a registration analysis between two adjacent scans.

[0226] In practice, the scanning range for each station is first determined based on the limit scanning distance and target tunnel size information set in step S105. To ensure data registration accuracy, there should be a certain overlap between two adjacent scanning positions to ensure that there are at least three common artificial targets as the basis for alignment between two scans, such as... Figure 10 As shown. Area of ​​the overlapping region. The calculation is as follows:

[0227]

[0228]

[0229]

[0230] in, It is the scanner's scanning distance. It is the complementary angle of the angle of incidence at the farthest target. It is the width of the tunnel surface. It is the distance between two scan positions. It is the length of the overlapping region.

[0231] S109. Based on the priority evaluation and division method in step S104, a genetic algorithm is used to optimize the optimal scanning scheme for point cloud data acquisition of the target tunnel, so as to capture the most complete scanned object with the fewest possible scanning stations while meeting the data quality requirements.

[0232] In practice, the target tunnel is first divided into regional categories, such as... Figure 11 As shown. Then set the objective function. This ensures that the tunnel point cloud is as complete as possible in different priority partitions.

[0233]

[0234] in, Indicates from the first Target points captured at each scanning location; This represents the total number of visible target points obtained from each selected scan location; It represents the total number of target points scanned.

[0235] Next, to prevent the optimization results from getting trapped in local optima, in addition to the aforementioned constraint on scan coverage, a constraint is imposed on the scan time. The maximum constraint value for the scan time is set... Let the scan time be set to the greedy algorithm's scan time within the same scene. Therefore, the optimization problem can be described as follows:

[0236]

[0237]

[0238] in, Indicates the first The second scan and the first +1 overlapping area between scans; This indicates the number of scan locations in the scan plan; It is the number of available scanning angular resolutions; This indicates the time required to move the scanner to the next location; Indicates the scanner at the The scan position is the first The time consumed during scanning at each scan resolution. Then, in the optimization process, chromosomes are generated using a combination of binary and integer encoding methods, and the SEGA algorithm is used for optimization. The convergence process is as follows: Figure 12 As shown, the optimized scanning scheme is as follows: Figure 13 As shown.

[0239] S110. Based on the target tunnel scanning optimization scheme obtained in step S109, point cloud data is collected using a land laser scanner, and then the point density analysis of the scan data is performed for verification.

[0240] In practice, firstly, based on the target tunnel scanning optimization scheme obtained in step 109, point cloud data of the tunnel is collected according to the scanning position and scanning resolution in the scheme. Then, the point clouds acquired from each station are registered one by one and integrated to generate the overall point cloud data of the target tunnel. Next, the local point density of the point cloud data is calculated to evaluate the data quality, using the following formula:

[0241]

[0242] in It is the number of points in the neighborhood of a point. It is the radius of the neighborhood of the nearest neighbor point at that point. The point density analysis results of the tunnel section point cloud data are as follows: Figure 14 As shown.

Claims

1. A method for optimizing point cloud scanning schemes in tunnel scenes considering regional priority assessment, characterized in that, include: 1) Based on the two-dimensional design drawings and three-dimensional BIM design model of the target tunnel scene, construct a three-dimensional solution space model of the target tunnel; 2) The three-dimensional solution space model of the target tunnel is divided into grids using the grid discretization function of BIM software, and the center point coordinates are calculated based on the divided grids to obtain the tunnel scanning object set; 3) Prioritize the sections of the target tunnel according to the relevant construction safety risk assessment guidelines and specifications; 4) Based on the set of tunnel scanning objects and the priority of the divided sections, a hierarchical scanning model for scanning sites is constructed; 5) Use a genetic algorithm to solve the hierarchical scanning model of the scanning sites to obtain the optimal scanning scheme; 6) Based on the optimal scanning scheme, the scanner is placed inside the target tunnel to collect point cloud data.

2. The method for optimizing a point cloud scanning scheme for a tunnel scene considering regional priority assessment as described in claim 1, characterized in that, The three-dimensional solution space model of the target tunnel includes the secondary lining inner wall, road surface, and drainage ditch cover.

3. The method for optimizing a point cloud scanning scheme for a tunnel scene considering regional priority assessment as described in claim 2, characterized in that, The steps for obtaining the tunnel scan object set are as follows: A1 uses the mesh discretization function of BIM software to divide the inner wall of the secondary lining into a mesh, and obtains the coordinates of the four vertices of the divided mesh. A2 calculates the coordinates of the center point of the grid based on the coordinates of the four vertices of the divided grid, and uses it as the target point for scanning; The coordinates of the center point of the grid are shown below: （1） In the formula, Indicates grid index, Indicates the first The coordinates of the center point of each grid; , , , The first The coordinates of the four vertices of the grid; A3 selects coordinate points at preset intervals on the center axis of the road surface as candidate scanning stations; A4 constructs a set of tunnel scanning objects based on the target scanning points and candidate scanning sites.

4. The method for optimizing a point cloud scanning scheme for a tunnel scene considering regional priority assessment as described in claim 1, characterized in that, The prioritization of the target tunnel sections is based on the risk scores for tunnel collapse accidents and water inrush and mudslide accidents, as shown below: （2） （3） （4） In the formula, This indicates the risk score for a tunnel collapse accident; This indicates the risk score for water inrush and mudslide accidents; Indicates the reduction factor; This indicates the score of the surrounding rock grade index; The score indicates the severity of fault rupture. The score indicates the permeability status index. This indicates the score for geological compliance index; This indicates the score for the construction method indicator; This indicates the score for the construction step distance index; The score represents the degree of karst development. This indicates the score of the fault rupture zone index; The scores represent the indicators of the surrounding water conditions; This indicates the score for each evaluation indicator.

5. The method for optimizing a point cloud scanning scheme for a tunnel scene considering regional priority assessment as described in claim 1, characterized in that, The objective function of the hierarchical scanning model for scanning sites is shown below: （5） In the formula, This represents the objective function of the hierarchical scanning model for scanning sites. Indicates the candidate scan site index; Indicates from the first The set of target points captured by each candidate scanning site; This represents the total number of target points captured from each candidate scan site; This indicates the total number of target points scanned.

6. The method for optimizing a point cloud scanning scheme for a tunnel scene considering regional priority assessment as described in claim 1, characterized in that, The constraints of the hierarchical scanning model at the scanning site include scanning distance constraints, occlusion constraints, quality constraints of the scanned object, level of detail constraints, and overlapping area constraints.

7. The method for optimizing a point cloud scanning scheme for a tunnel scene considering regional priority assessment as described in claim 6, characterized in that, The scanning distance constraint is as follows: （6） （7） In the formula, Indicates the scanning distance from the candidate scanning site to the target scanning point; Indicates the scanning radius of the scanner; Indicates the coordinates of the target point being scanned; Indicates the coordinates of the candidate scan sites; The occlusion constraint adds a candidate scanning station to the area where scanning is obstructed within the tunnel. The area where the scan was blocked included the emergency parking lane; The quality constraints of the scanned object are as follows: （8） （9） In the formula, Indicates the angle of incidence of the target point being scanned; This represents the line-of-sight vector from the target point to the candidate scanning site; This represents the normal vector of the target point being scanned; Represents the norm; Indicates the limit scanning angle; The level-of-detail constraints are as follows: （10） （11） （12） In the formula, This represents the horizontal distance between two adjacent scanned target points; Indicates the horizontal angle of incidence; This indicates the angular resolution; different priority segments have different angular resolutions. Indicates the distance between the candidate scanning site and the target scanning point; Indicates the vertical distance between two adjacent scan target points; Indicates the vertical angle of incidence; Indicates the first The set of target points captured by each candidate scanning site; This indicates the level of detail required for horizontal scanning; This indicates the level of detail required for vertical scanning; Indicates the coordinates of the target point being scanned; The area constraint of the overlapping region is as follows: （13） （14） （15） In the formula, Indicates the scanning radius of the scanner; Indicates the width of the road surface; The complementary angles represent the incident angles at the farthest scanned target point; Indicates the distance between two candidate scan sites; Indicates the length of the overlapping region; This indicates the area of ​​the overlapping region.

8. The method for optimizing a point cloud scanning scheme for a tunnel scene considering regional priority assessment as described in claim 1, characterized in that, When using a genetic algorithm to solve the hierarchical scanning model for scanning sites, a scanning time constraint is also introduced. The hierarchical scanning model for scanning sites is then shown below: （16） （17） In the formula, Represents the maximization function; This represents the objective function of the hierarchical scanning model for scanning sites. Indicates scan time; This represents the maximum constraint value for scan time; Indicates the candidate scan site index; Indicates the total number of candidate scanning sites; Indicates the first The candidate scanning sites and the first Overlapping area between candidate scanning sites; This indicates the time it takes to move the scanner to the next candidate scanning site; Indicates the scan resolution index; Indicates the total number of scan resolutions; Indicates the scanner at the The candidate scanning site was the first The time consumed when performing a scan at each scan resolution.

9. The method for optimizing a point cloud scanning scheme for a tunnel scene considering regional priority assessment as described in claim 1, characterized in that, The scanner includes a laser scanner.

10. The method for optimizing a point cloud scanning scheme for a tunnel scene considering regional priority assessment as described in claim 1, characterized in that, Point cloud data acquired using scanners deployed according to an optimal scanning scheme is evaluated for quality based on local point density, as shown below: （18） In the formula, Represents local point density; This represents the number of points in the neighborhood of the point cloud data; This represents the neighborhood radius of the point cloud data to the nearest neighbor.

Images