Method, device and equipment for processing point cloud data of tunnel face, medium and product

Abstract

The invention relates to the technical field of point cloud data processing, and discloses a face point cloud data processing method, device, equipment, medium and product, and the face point cloud data processing method comprises the steps: carrying out the plane fitting of the face point cloud data, so as to extract a main plane where a face is located; projecting the tunnel face point cloud data to the main plane to obtain a projection result; performing density identification on the projection result on the main plane to obtain a density distribution condition of the tunnel face point cloud data corresponding to the projection result; the tunnel face point cloud data is screened according to the density distribution condition to obtain the first target point cloud data, the tunnel face point cloud data are screened according to the density distribution condition, non-tunnel face point cloud data in the tunnel face point cloud data are reduced, and the accuracy of the first target point cloud data is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of point cloud data processing, and in particular to a method, device, equipment, medium and product for processing tunnel face point cloud data. Background Art

[0002] In underground projects (such as tunnel projects, underground cavern projects, etc.), the face refers to the working face that is continuously excavated during the construction of underground projects. The face is the forefront of underground construction and is also the exposed surface of newly excavated rock or soil. For example, in tunnel construction, the face is the face of the end of the tunnel being excavated by the shield machine. In the construction and monitoring of underground projects, face point cloud data, as an important source of three-dimensional information, is widely used in geological exploration, tunnel construction, structural monitoring and other fields.

[0003] In related technologies, face point cloud data is usually obtained through laser scanning, optical imaging and other technologies, which can effectively reflect the geometric characteristics and spatial distribution of underground structures. However, in the actual acquisition process, due to environmental factors, equipment errors and surrounding geological conditions, face point cloud data often contains a lot of noise, non-face data and missing data. The existence of a large amount of noise, non-face data and missing data will not only lead to inaccurate data analysis, but also may interfere with subsequent engineering decisions and implementation. Summary of the invention

[0004] In view of this, the present invention provides a method, device, equipment, medium and product for processing tunnel face point cloud data to solve the problem of inaccurate tunnel face point cloud data.

[0005] In a first aspect, the present invention provides a method for processing palm face point cloud data, comprising: performing plane fitting on the palm face point cloud data to extract the main plane where the palm face is located; projecting the palm face point cloud data onto the main plane to obtain a projection result; performing density identification on the projection result on the main plane to obtain the density distribution of the palm face point cloud data corresponding to the projection result; and screening the palm face point cloud data according to the density distribution to obtain first target point cloud data.

[0006] The present invention performs plane fitting on the palm face point cloud data to extract the main plane where the palm face is located, and provides a reference plane for subsequent processing, so as to facilitate the subsequent screening of the palm face point cloud data. The present invention projects the palm face point cloud data onto the main plane to obtain a projection result. The present invention converts three-dimensional data into two-dimensional data through projection, thereby reducing the complexity of the palm face point cloud data. The present invention performs density identification on the projection result on the main plane to obtain the density distribution of the palm face point cloud data corresponding to the projection result, and screens the palm face point cloud data according to the density distribution to obtain the first target point cloud data. In the present invention, the tunnel face point cloud data may include point cloud data corresponding to non-tunnel faces. The main plane where the tunnel face is located is perpendicular to the point cloud data corresponding to the tunnel face. After being projected onto the main plane where the tunnel face is located, the density is relatively sparse due to the small number of overlapping point cloud data. After the tunnel face point cloud data is projected onto the main plane, the point cloud data corresponding to the non-tunnel faces densely overlap at the tunnel contour on the main plane where the tunnel face is located. Therefore, the density of the point cloud data corresponding to the non-tunnel faces is relatively dense after being projected onto the main plane. Therefore, the tunnel face point cloud data is screened according to the density distribution, the point cloud data corresponding to the tunnel face in the tunnel face point cloud data is retained, and the point cloud data corresponding to the non-tunnel face is eliminated, so as to improve the accuracy of the first target point cloud data and provide a reliable data basis for the subsequent analysis and processing of the tunnel face point cloud data.

[0007] In an optional embodiment, the method for processing the point cloud data of the face further includes: determining multiple local normal vectors of the first target point cloud data, adjusting the directions of the multiple local normal vectors for consistency, and obtaining a normal field; constructing a structural model according to the normal field, uniformly sampling the structural model, and obtaining second target point cloud data; wherein the structural model is used to describe the geometric structure of the first target point cloud data.

[0008] The present invention determines multiple local normal vectors of the first target point cloud data and adjusts their directions to obtain a normal field, so that the first target point cloud data can have higher consistency and standardization in the description of the geometric structure, avoiding errors or errors caused by problems such as inconsistent directions of the first target point cloud data. The present invention constructs a structural model based on the normal field, uniformly samples the structural model, obtains the second target point cloud data, effectively eliminates the noise in the first target point cloud data, and retains the precise geometric information of the face, providing high-quality point cloud data for subsequent construction analysis and processing.

[0009] In an optional embodiment, plane fitting is performed on the point cloud data of the tunnel face to extract the main plane where the tunnel face is located, including: identifying a point cloud data set that conforms to a preset plane model in the point cloud data of the tunnel face; wherein the preset plane model is an abstract representation model that describes the plane using equations; randomly selecting a preset number of random point cloud data from the point cloud data set to obtain coordinates of the random point cloud data; determining normal vector parameters of the preset plane model according to the coordinates; iteratively fitting the coordinates and the normal vector parameters to obtain a target plane model; and extracting the main plane where the tunnel face is located in the target plane model.

[0010] In an optional implementation, density identification is performed on the projection result on the main plane to obtain the density distribution of the point cloud data of the face corresponding to the projection result, including: determining the neighboring points of each point cloud data in the face point cloud data corresponding to the projection result on the main plane; the neighboring points are the points closest to the point cloud data; determining the distance between each point cloud data and the neighboring points of each point cloud data; generating an average resolution according to the distance and the number of face point cloud data, and obtaining the minimum density point and the neighborhood radius according to the average resolution; determining the neighborhood set of each point cloud data according to the neighborhood radius; determining the core point according to the neighborhood set and the minimum density point; adding the core point to the cluster grouping, judging whether each point cloud data in the neighborhood set can be added to the cluster grouping, so as to obtain the density distribution according to the multiple point cloud data that can be added to the cluster grouping.

[0011] In an optional implementation, the palm face point cloud data is screened according to density distribution to obtain first target point cloud data, including: retaining first palm face point cloud data having a first density distribution, discarding second palm face point cloud data having a second density distribution, and using the retained first palm face point cloud data as the first target point cloud data; wherein the density corresponding to the first distribution is less than the density corresponding to the second distribution.

[0012] In an optional implementation, constructing a structural model according to a normal field includes: constructing a Poisson's equation according to the normal field, and obtaining structural model parameters by solving the Poisson's equation; and constructing the structural model according to the structural model parameters.

[0013] In a second aspect, the present invention provides a processing device for palm face point cloud data, comprising: a main plane extraction module, used to perform plane fitting on the palm face point cloud data to extract the main plane where the palm face is located; a projection module, used to project the palm face point cloud data onto the main plane to obtain a projection result; a density distribution determination module, used to perform density identification on the projection result on the main plane to obtain the density distribution of the palm face point cloud data corresponding to the projection result; a point cloud data screening module, used to screen the palm face point cloud data according to the density distribution to obtain the first target point cloud data.

[0014] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method for processing face point cloud data of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for processing face point cloud data of the first aspect or any corresponding embodiment thereof.

[0016] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions, wherein the computer instructions are used to enable a computer to execute the method for processing tunnel face point cloud data of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the related technologies, the drawings required for use in the specific embodiments or the related technical descriptions will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 4 is a flow chart of a method for processing tunnel face point cloud data according to an embodiment of the present invention.

[0019] Figure 2 4 is a schematic diagram of an execution flow of a density-based spatial clustering algorithm according to an embodiment of the present invention.

[0020] FIG. 3( a ) is a schematic diagram of a tunnel face model before tunnel face point cloud data screening according to an embodiment of the present invention.

[0021] FIG3( b ) is a schematic diagram of a tunnel face model after tunnel face point cloud data screening according to an embodiment of the present invention.

[0022] Figure 4 4 is a flow chart of another method for processing tunnel face point cloud data according to an embodiment of the present invention.

[0023] FIG. 5( a ) is a schematic diagram of a dodecahedron before noise reduction according to an embodiment of the present invention.

[0024] FIG5( b ) is a schematic diagram of a dodecahedron after noise reduction according to an embodiment of the present invention.

[0025] FIG6( a ) is a schematic diagram of tunnel face point cloud data before noise reduction according to an embodiment of the present invention.

[0026] FIG6( b ) is a schematic diagram of the tunnel face point cloud data after noise reduction according to an embodiment of the present invention.

[0027] Figure 7 4 is a structural block diagram of a device for processing tunnel face point cloud data according to an embodiment of the present invention.

[0028] Figure 8 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0030] In the construction and monitoring of underground cavern projects, face point cloud data, as an important source of three-dimensional information, is widely used in geological exploration, tunnel construction, structural monitoring and other fields. Face point cloud data is usually obtained through laser scanning, optical imaging and other technologies, which can effectively reflect the geometric characteristics and spatial distribution of underground structures. However, in the actual acquisition process, due to environmental factors, equipment errors and surrounding geological conditions, face point cloud data often contains a lot of noise and missing data. The presence of noise not only leads to inaccurate data analysis, but also may interfere with subsequent engineering decisions and implementation. For example, in the design stage of the tunnel, if unprocessed face point cloud data is used for model construction, it may lead to design defects, thus affecting construction safety and engineering quality. In addition, noise can also make data visualization poor.

[0031] An embodiment of the present invention provides a method for processing tunnel face point cloud data, which screens tunnel face point cloud data according to density distribution and reduces non-tunnel face point cloud data in the tunnel face point cloud data, so as to improve the accuracy of the first target point cloud data.

[0032] According to an embodiment of the present invention, an embodiment of a method for processing tunnel face point cloud data is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0033] In this embodiment, a method for processing tunnel face point cloud data is provided, which can be used in computer equipment. Figure 1 is a flow chart of a method for processing tunnel face point cloud data according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0034] Step S101, plane fitting is performed on the tunnel face point cloud data to extract the main plane where the tunnel face is located.

[0035] The tunnel face point cloud data is a set of three-dimensional coordinate data of a large number of discrete points on the tunnel face obtained through measurement technology. Specifically, the measurement technology can be laser scanning, optical imaging and other technologies.

[0036] In some optional embodiments, the RANSAC (Random Sample Consensus) algorithm is used to perform plane fitting on the tunnel face point cloud data to extract the main plane where the tunnel face is located. The RANSAC algorithm can quickly identify the point set that best fits the plane model in the tunnel face point cloud data and eliminate outliers. The principle of the RANSAC algorithm is to identify the point set that best fits the preset plane model through iterative sampling and fitting.

[0037] Specifically, the execution process of the RANSAC algorithm includes: identifying a point cloud data set that conforms to a preset plane model in the tunnel face point cloud data; wherein the preset plane model is an abstract representation model that describes a plane using an equation; randomly selecting a preset number of random point cloud data from the point cloud data set to obtain the coordinates of the random point cloud data; determining the normal vector parameters of the preset plane model based on the coordinates; iteratively fitting the coordinates and normal vector parameters to obtain a target plane model; and extracting the main plane where the tunnel face is located in the target plane model.

[0038] For the preset planar model, the fitting equation is:

[0039] Ax+By+Cz+D=0,

[0040] Among them, (x, y, z) is the point coordinates in the face point cloud data, A, B, C are the normal vectors of the preset plane, and D is the plane constant.

[0041] Step S102, projecting the tunnel face point cloud data onto the main plane to obtain a projection result.

[0042] Among them, the tunnel face point cloud data includes not only the point cloud data corresponding to the tunnel face, but also some point cloud data corresponding to non-tunnel faces, such as the point cloud data corresponding to the tunnel side wall and the point cloud data corresponding to the tunnel arch edge.

[0043] In some optional embodiments, the main plane where the tunnel face is located tends to be perpendicular to the central axis area of ​​the tunnel, although not completely perpendicular, but the angle formed by the main plane and the central axis is large, and the curved surface around the tunnel face tends to be parallel to the central axis of the tunnel, although not completely parallel, but the angle formed by the curved surface around the tunnel face and the central axis of the tunnel is small. Therefore, the tunnel face point cloud data is projected onto the main plane that tends to be perpendicular to the central axis of the tunnel to observe the density distribution of the tunnel face point cloud data.

[0044] In some optional implementations, the formula used to project the tunnel face point cloud data onto the main plane is:

[0045] p proj =p i -(n·p i )n,

[0046] Among them, p i is any point in the point cloud data of the face, n is the normal vector of the principal plane, p is proj is the point cloud coordinate after projection.

[0047] In some optional embodiments, the tunnel face point cloud data is projected onto the main plane. After the projection result is obtained, the point cloud data corresponding to the tunnel face is perpendicular to the main plane, and there are fewer points of overlap in the projection, so the density is relatively sparse, while the point cloud data corresponding to the non-tunnel face densely overlaps at the tunnel contour position on the main plane, so the density is relatively dense.

[0048] Step S103: Density identification is performed on the projection result on the main plane to obtain density distribution of the tunnel face point cloud data corresponding to the projection result.

[0049] In some optional embodiments, DBSCAN (Density Based Spatial Clustering of Applications with Noise) is used to identify high-density areas. The principle of the DBSCAN algorithm is to cluster dense areas by finding a set of points with higher density (i.e., the distance between a point and its neighbors is less than a certain threshold, and the number of points in the neighborhood exceeds a certain minimum value).

[0050] Specifically, the execution process of the DBSCAN algorithm includes: determining the neighboring points of each point cloud data in the face point cloud data corresponding to the projection result on the main plane; the neighboring points are the points closest to the point cloud data; determining the distance between each point cloud data and the neighboring points of each point cloud data; generating an average resolution based on the distance and the number of face point cloud data, and obtaining the minimum density point and neighborhood radius based on the average resolution; determining the neighborhood set of each point cloud data based on the neighborhood radius; determining the core point based on the neighborhood set and the minimum density point; adding the core point to the cluster grouping, judging whether each point cloud data in the neighborhood set can be added to the cluster grouping, so as to obtain the density distribution based on multiple point cloud data that can be added to the cluster grouping.

[0051] The average resolution refers to the average distance between adjacent point clouds in the point cloud data. By counting the neighborhood relationship of all points, the spatial resolution of the tunnel face point cloud data can be estimated. The calculation principle of the average resolution is as follows: first, a KD tree (K-Dimensional Tree) is constructed, and the KD tree is used to accelerate the search of the tunnel face point cloud data to obtain each point p i The nearest k neighbor points, the distance between each point and its neighbor point is recorded as d i , then for all points p i , calculate the distance d between it and the nearest neighbor i The average value is the average resolution res. The formula for calculating the average resolution res is:

[0052]

[0053] Among them, res is the average resolution, N is the total number of face point cloud data, d i For each point p i The distance between its neighbors.

[0054] In some optional implementations, the neighborhood radius R reflects the local neighborhood range around each point. In order to ensure that there are enough neighborhood points in the tunnel face point cloud data of different resolutions, the neighborhood radius is usually set to twice the average resolution, that is, the calculation formula of the neighborhood radius R is:

[0055] R = 2·res,

[0056] Among them, R is the neighborhood radius and res is the average resolution.

[0057] In some optional implementations, the minimum density point N represents the number of points that must be included in each cluster at least. The minimum density point N is usually a function of the average resolution and can be set according to actual needs. A common setting is to set it according to the empirical value of the point cloud density, for example:

[0058]

[0059] Among them, N is the minimum density point, res is the average resolution, and α is an empirical constant. The specific value of the empirical constant can be adjusted according to the properties of the face point cloud data. This formula indicates that in point cloud data with lower resolution, the number of points in the neighborhood is required to be less, while in point cloud data with higher resolution, the number of points in the neighborhood is required to be more.

[0060] For example, Figure 2 As shown in the figure, it is a schematic diagram of the execution process of the density-based spatial clustering algorithm. Two parameters are input: the minimum density point N and the neighborhood radius R. An unprocessed point i is selected, with point i as the center and the neighborhood radius R as the radius. The number M of point clouds in the circle is calculated, and it is determined whether the number M of point clouds in the circle is greater than the minimum density point N. When the number M of point clouds in the circle is less than or equal to the minimum density point N, point i is determined as an edge point or a noise point, and marked as processed, and the step of selecting unprocessed point i is returned. When the number M of point clouds in the circle is greater than the minimum density point N, point i is determined as a core point, and objects whose density can be reached by point i are found from all points to form a point cloud set. It is determined whether all points have been processed. If they have been processed, the task is terminated. If they have not been processed, the step of selecting unprocessed point i is returned again until all points are processed.

[0061] In the embodiment of the present invention, by calculating the average resolution and automatically adjusting the parameters of DBSCAN (minimum density point N and neighborhood radius R), it is possible to adapt to different types of point cloud data and ensure that the tunnel face area and its surrounding area can be effectively separated in various situations.

[0062] Step S104: screening the tunnel face point cloud data according to the density distribution to obtain first target point cloud data.

[0063] In some optional implementations, the palm face point cloud data are screened according to density distribution to obtain first target point cloud data, including: retaining first palm face point cloud data having a first density distribution, discarding second palm face point cloud data having a second density distribution, and using the retained first palm face point cloud data as the first target point cloud data; wherein the density corresponding to the first distribution is less than the density corresponding to the second distribution.

[0064] As shown in Figure 3(a), a schematic diagram of a tunnel face model before the tunnel face point cloud data is screened according to an embodiment of the present invention, point cloud data of different densities exist on the front and side of the tunnel face model. Figure 3(b) is a schematic diagram of a tunnel face model after the tunnel face point cloud data is screened according to an embodiment of the present invention. After screening by the embodiment of the present invention, the point cloud data density on the front and side of the tunnel face model is uniform, and there is no scattered point cloud data.

[0065] The method for processing palm face point cloud data provided in this embodiment performs plane fitting on the palm face point cloud data to extract the main plane where the palm face is located, and provides a reference plane for subsequent processing, so as to facilitate the subsequent screening of palm face point cloud data. The embodiment of the present invention projects the palm face point cloud data onto the main plane to obtain a projection result. The present invention converts three-dimensional data into two-dimensional data through projection, thereby reducing the complexity of the palm face point cloud data. The embodiment of the present invention performs density recognition on the projection result on the main plane to obtain the density distribution of the palm face point cloud data corresponding to the projection result, and screens the palm face point cloud data according to the density distribution to obtain the first target point cloud data. In an embodiment of the present invention, the tunnel face point cloud data may include point cloud data corresponding to non-tunnel faces. The main plane where the tunnel face is located is perpendicular to the point cloud data corresponding to the tunnel face. After being projected onto the main plane where the tunnel face is located, the density is relatively sparse due to the small number of overlapping point cloud data. After the tunnel face point cloud data is projected onto the main plane, the point cloud data corresponding to the non-tunnel faces densely overlap at the tunnel contour on the main plane where the tunnel face is located. Therefore, the density of the point cloud data corresponding to the non-tunnel faces is relatively dense after being projected onto the main plane. Therefore, the tunnel face point cloud data is screened according to the density distribution, the point cloud data corresponding to the tunnel face in the tunnel face point cloud data is retained, and the point cloud data corresponding to the non-tunnel face is eliminated, so as to improve the accuracy of the first target point cloud data and provide a reliable data basis for the subsequent analysis and processing of the tunnel face point cloud data.

[0066] In this embodiment, a method for processing tunnel face point cloud data is provided, which can be used in computer equipment. Figure 4 FIG. 4 is a flowchart of another method for processing tunnel face point cloud data according to an embodiment of the present invention. Figure 4 As shown, the process includes the following steps:

[0067] Step S401, plane fitting is performed on the tunnel face point cloud data to extract the main plane where the tunnel face is located. Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0068] Step S402, projecting the tunnel face point cloud data onto the main plane to obtain a projection result.

[0069] Step S403: Density identification is performed on the projection result on the main plane to obtain density distribution of the tunnel face point cloud data corresponding to the projection result.

[0070] Step S404: screening the tunnel face point cloud data according to the density distribution to obtain first target point cloud data.

[0071] Step S405 , determining a plurality of local normal vectors of the first target point cloud data, and adjusting the directions of the plurality of local normal vectors for consistency to obtain a normal field.

[0072] In some optional implementations, the process of determining multiple local normal vectors of the first target point cloud data is: using a KD tree to perform an accelerated search on the first target point cloud data, finding the neighborhood of each point, and for each point p i , by querying its neighborhood point set Calculate the local normal vector n i The local normal vector n i It can be obtained by performing eigenvalue decomposition on the covariance matrix Q:

[0073]

[0074] Among them, Q is the covariance matrix, q j is any point p i The jth nearest neighbor of For any point p i The neighborhood point set, k is the number of neighboring points, λ i is the eigenvalue of the covariance matrix Q, is the eigenvector corresponding to the eigenvalue, is the feature vector The transpose of .

[0075] In some optional implementations, let λ1>λ2>λ3 be the eigenvalues ​​of the covariance matrix Q, is the eigenvector corresponding to the eigenvalue λ1, is the eigenvector corresponding to the eigenvalue λ2, is the eigenvector corresponding to the eigenvalue λ3, then the local normal vector n i for:

[0076]

[0077] Among them, n i is the local normal vector, is the eigenvector corresponding to the minimum eigenvalue λ3.

[0078] In some optional implementations, the process of adjusting the directions of multiple local normal vectors to be consistent is as follows: a reference point p0 is selected from the first target point cloud data, and its local normal vector direction n0 is considered to be the positive direction, and for its neighboring point p i The local normal vector n i , by judging n i The sign of the dot product with n0 determines whether to invert n i , the specific formula is:

[0079]

[0080] Among them, n′ i is the final local normal vector, n i is the local normal vector of the original direction, -n i is the local normal vector in the opposite direction, if n i n0>0, the final local normal vector direction remains unchanged. If n i n0≤0, the final local normal vector has the opposite direction.

[0081] Step S406, constructing a structural model according to the normal field, uniformly sampling the structural model, and obtaining second target point cloud data; wherein the structural model is used to describe the geometric structure of the first target point cloud data.

[0082] In some optional implementations, constructing a structural model according to a normal field includes: constructing a Poisson's equation according to the normal field, and obtaining structural model parameters by solving the Poisson's equation; and constructing a structural model according to the structural model parameters.

[0083] The mathematical form of Poisson's equation is:

[0084]

[0085] Among them, Δf is the implicit function to be reconstructed, is the Hamiltonian operator, N is the normal field, Represents the divergence of the normal field.

[0086] In some optional embodiments, the implicit function is solved by minimizing the following energy function:

[0087]

[0088] Where E(f) is the energy function, Ω is the normal field, is the implicit function to be reconstructed, N is the normal field, and V is the position function of the point cloud data.

[0089] In an embodiment of the present invention, by extracting the zero isosurface in the implicit function, the structural model parameters are obtained, and the structural model is constructed according to the structural model parameters. Exemplarily, the structural model may be a triangular mesh model, and by extracting the zero isosurface in the implicit function, the coordinates of three points of the triangular mesh model are obtained, and the triangular mesh model is constructed according to the coordinates of the three points.

[0090] In some optional implementations, the formula for uniformly sampling the structural model is:

[0091] g i =αv1+βv2+(1-α-β)v3,

[0092] Among them, g i is a sampled point in a small mesh triangle in the triangular mesh model, v1, v2, v3 are the vertices of the mesh triangle, α is a random number, β is a random number, α and β satisfy 0≤α+β≤1.

[0093] In an embodiment of the present invention, by sampling from a grid model, high-precision point cloud data can be obtained, while reducing noise interference and effectively retaining the detail information of the face. For example, FIG5(a) is a schematic diagram of a dodecahedron before denoising according to an embodiment of the present invention. The point cloud on the dodecahedron is relatively dense, with a strong sense of grain, the surface of the object looks relatively rough, and the point distribution in the details is relatively messy; FIG5(b) is a schematic diagram of a dodecahedron after denoising according to an embodiment of the present invention. After denoising processing according to an embodiment of the present invention, the point cloud on the dodecahedron in FIG5(b) is relatively sparse, with a weak sense of grain, the surface of the object appears smoother and regular, the point distribution in the details is more uniform, the shape and outline of the object can be seen more clearly, and it is visually neater and clearer. FIG6(a) is a schematic diagram of the point cloud data of the face before denoising according to an embodiment of the present invention. The point cloud data of the face before denoising shows a relatively obvious linear texture. The point cloud data looks relatively dense and has a strong sense of grain, and contains more noise and redundant information. FIG6(b) is a schematic diagram of the point cloud data of the face after denoising according to an embodiment of the present invention. The surface texture of the point cloud data of the face after denoising is different from that of the original data. The linear texture becomes less obvious, and the overall texture is smoother and more uniform. Some small noise points are removed, the point cloud data is relatively sparse, the sense of grain is weakened, the data is more concise, and the main shape and structural characteristics of the rock are highlighted, making it easier to observe and analyze the macroscopic morphology and main surface characteristics of the rock.

[0094] The method for processing the face point cloud data provided in this embodiment determines multiple local normal vectors of the first target point cloud data and adjusts their directions to obtain a normal field, which can make the first target point cloud data more consistent and standardized in the description of the geometric structure, and avoid errors or errors caused by problems such as inconsistent directions of the first target point cloud data. The present invention constructs a structural model based on the normal field, uniformly samples the structural model, obtains the second target point cloud data, effectively eliminates the noise in the first target point cloud data, and retains the precise geometric information of the face, providing high-quality point cloud data for subsequent construction analysis and processing.

[0095] In this embodiment, a processing device for face point cloud data is also provided, which is used to implement the above embodiments and preferred implementations, and will not be repeated for what has been described. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0096] This embodiment provides a device for processing face point cloud data, such as Figure 7 As shown, including:

[0097] The main plane extraction module 701 is used to perform plane fitting on the tunnel face point cloud data to extract the main plane where the tunnel face is located.

[0098] The projection module 702 is used to project the point cloud data of the tunnel face onto the main plane to obtain a projection result.

[0099] The density distribution determination module 703 is used to perform density identification on the projection result on the main plane to obtain the density distribution of the tunnel face point cloud data corresponding to the projection result.

[0100] The point cloud data screening module 704 is used to screen the tunnel face point cloud data according to the density distribution to obtain the first target point cloud data.

[0101] In some optional implementations, the device for processing the tunnel face point cloud data further includes:

[0102] The normal field determination module is used to determine multiple local normal vectors of the first target point cloud data, and adjust the directions of the multiple local normal vectors to be consistent to obtain a normal field.

[0103] The uniform extraction module is used to construct a structural model according to the normal field, and uniformly sample the structural model to obtain the second target point cloud data; wherein the structural model is used to describe the geometric structure of the first target point cloud data.

[0104] In some optional implementations, the main plane extraction module 701 includes:

[0105] The point cloud data recognition unit is used to identify a point cloud data set that conforms to a preset plane model in the tunnel face point cloud data; wherein the preset plane model is an abstract representation model that describes a plane using equations.

[0106] The point cloud data random selection unit is used to randomly select a preset number of random point cloud data from the point cloud data set and obtain the coordinates of the random point cloud data.

[0107] The normal vector parameter determination unit is used to determine the normal vector parameters of the preset plane model according to the coordinates.

[0108] The iterative fitting unit is used to iteratively fit the coordinate and normal vector parameters to obtain the target plane model.

[0109] The main plane extraction unit is used to extract the main plane where the tunnel face is located in the target plane model.

[0110] In some optional implementations, the density distribution determination module 703 includes:

[0111] The neighbor point determination unit is used to determine the neighbor points of each point cloud data in the tunnel face point cloud data corresponding to the projection result on the main plane; the neighbor point is the point closest to the point cloud data.

[0112] The distance determination unit is used to determine the distance between each point cloud data and its neighboring points.

[0113] The parameter determination unit is used to generate an average resolution according to the distance and the number of the tunnel face point cloud data, and obtain the minimum density point and the neighborhood radius according to the average resolution.

[0114] The core point determination unit is used to determine the neighborhood set of each point cloud data according to the neighborhood radius; and determine the core point according to the neighborhood set and the minimum density point.

[0115] The density distribution determination unit is used to add the core point to the cluster group, determine whether each point cloud data in the neighborhood set can be added to the cluster group, and obtain the density distribution according to the multiple point cloud data that can be added to the cluster group.

[0116] In some optional implementations, the point cloud data screening module 704 includes:

[0117] The first target point cloud data determination unit is used to retain the first palm face point cloud data whose density distribution is the first distribution, discard the second palm face point cloud data whose density distribution is the second distribution, and use the retained first palm face point cloud data as the first target point cloud data; wherein the density corresponding to the first distribution is less than the density corresponding to the second distribution.

[0118] In some optional embodiments, the uniform extraction module includes:

[0119] The Poisson equation solving unit is used to construct the Poisson equation according to the normal field and obtain the structural model parameters by solving the Poisson equation.

[0120] The structural model building unit is used to build the structural model according to the structural model parameters.

[0121] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0122] The processing device for the tunnel face point cloud data in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0123] The embodiment of the present invention also provides a computer device having the above Figure 7 The device for processing the tunnel face point cloud data is shown.

[0124] See also Figure 8 , Figure 8 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 8 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 8 A processor 10 is taken as an example.

[0125] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0126] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0127] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0128] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0129] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0130] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0131] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0132] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for processing tunnel face point cloud data, characterized in that: The method comprises: Perform plane fitting on the tunnel face point cloud data to extract the main plane where the tunnel face is located; Projecting the tunnel face point cloud data onto the main plane to obtain a projection result; Performing density identification on the projection result on the main plane to obtain density distribution of the tunnel face point cloud data corresponding to the projection result; The tunnel face point cloud data is screened according to the density distribution to obtain first target point cloud data.

2. The method according to claim 1, characterized in that The method further comprises: Determine a plurality of local normal vectors of the first target point cloud data, and adjust the directions of the plurality of local normal vectors for consistency to obtain a normal field; A structural model is constructed according to the normal field, and the structural model is uniformly sampled to obtain second target point cloud data; wherein the structural model is used to describe the geometric structure of the first target point cloud data.

3. The method according to claim 1 or 2, characterized in that: The performing plane fitting on the tunnel face point cloud data to extract the main plane where the tunnel face is located includes: Identifying a point cloud data set that conforms to a preset plane model in the tunnel face point cloud data; wherein the preset plane model is an abstract representation model that describes a plane using equations; Randomly selecting a preset number of random point cloud data from the point cloud data set, and obtaining coordinates of the random point cloud data; Determine the normal vector parameters of the preset plane model according to the coordinates; Iteratively fitting the coordinates and the normal vector parameters to obtain a target plane model; In the target plane model, the main plane where the tunnel face is located is extracted.

4. The method according to claim 1 or 2, characterized in that: The performing density identification on the projection result on the main plane to obtain the density distribution of the tunnel face point cloud data corresponding to the projection result includes: Determine the neighboring point of each point cloud data in the tunnel face point cloud data corresponding to the projection result on the main plane; the neighboring point is the point closest to the point cloud data; Determine the distance between each point cloud data and its neighboring points; Generate an average resolution according to the distance and the amount of the tunnel face point cloud data, and obtain a minimum density point and a neighborhood radius according to the average resolution; Determine a neighborhood set of each point cloud data according to the neighborhood radius; determine a core point according to the neighborhood set and the minimum density point; The core point is added to the cluster group, and it is determined whether each point cloud data in the neighborhood set can be added to the cluster group, so as to obtain the density distribution according to the multiple point cloud data that can be added to the cluster group.

5. The method according to claim 1 or 2, characterized in that: The screening of the tunnel face point cloud data according to the density distribution to obtain first target point cloud data includes: The first palm face point cloud data whose density distribution is the first distribution is retained, the second palm face point cloud data whose density distribution is the second distribution is discarded, and the retained first palm face point cloud data is used as the first target point cloud data; wherein the density corresponding to the first distribution is less than the density corresponding to the second distribution.

6. The method according to claim 2, characterized in that The step of constructing a structural model according to the normal field comprises: Constructing a Poisson equation according to the normal field, and obtaining structural model parameters by solving the Poisson equation; The structural model is constructed according to the structural model parameters.

7. A processing device for face point cloud data, characterized in that: The device comprises: The main plane extraction module is used to perform plane fitting on the tunnel face point cloud data to extract the main plane where the tunnel face is located; A projection module, used for projecting the tunnel face point cloud data onto the main plane to obtain a projection result; A density distribution determination module is used to perform density identification on the projection result on the main plane to obtain the density distribution of the tunnel face point cloud data corresponding to the projection result; The point cloud data screening module is used to screen the tunnel face point cloud data according to the density distribution to obtain first target point cloud data.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for processing the face point cloud data according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method for processing tunnel face point cloud data according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions, wherein the computer instructions are used to enable a computer to execute the method for processing tunnel face point cloud data according to any one of claims 1 to 6.

Images