Rock mass trace extraction method and device, electronic equipment and storage medium

Abstract

The invention provides a rock mass trace extraction method and device, electronic equipment and a storage medium, and relates to the technical field of rock mass monitoring. Three-dimensional point cloud data of a target rock mass is acquired; extracting edge points of a target rock mass in the three-dimensional point cloud data to obtain an initial shrinkage feature point set; extracting L1 median points in the initial shrinkage feature point set by adopting an L1 median algorithm to obtain skeleton point cloud data; determining a neighborhood point set of each skeleton point in the skeleton point cloud data to obtain a plurality of skeleton point neighborhood sets; constructing an undirected graph according to each skeleton point neighborhood set, and determining a topological relation among skeleton points in each skeleton point neighborhood set according to the undirected graph; generating a trace of the target rock mass according to the topological relation between the skeleton points in the skeleton point neighborhood set; therefore, the accurate trace of the target rock mass is efficiently and automatically extracted.

Description

Technical Field

[0001] The present invention relates to the technical field of rock mass monitoring, and more particularly, to a method, device, electronic device, and storage medium for extracting rock mass trace lines. Background Art

[0002] The trace lines of the structural planes of a rock mass are generated by the intersection of the rock mass surface and the rock discontinuity planes, and are used to characterize the basic parameter quality of the rock extension, providing important reference indicators in the rock mass quality classification and the analysis of the structural stability of the rock mass.

[0003] Currently, the measurement of the exposed trace lines of the structural planes of a rock mass is carried out on-site by hand-held devices. For example, manual measurements are performed by operators holding devices such as geological compasses, clinometers, tape measures, and roughness profile meters. This manual measurement method relies on the experience of the operators, has a low accuracy rate, and is inefficient. Summary of the Invention

[0004] The problem solved by the present invention is how to improve the extraction efficiency and accuracy of rock mass trace lines.

[0005] To solve the above problems, the present invention provides a method, device, electronic device, and storage medium for extracting rock mass trace lines.

[0006] In a first aspect, the present invention provides a method for extracting rock mass trace lines, including: Obtaining three-dimensional point cloud data of a target rock mass; Extracting the edge points of the target rock mass from the three-dimensional point cloud data to obtain an initial set of shrinkage feature points; Using the L1 median algorithm to extract the L1 median points from the initial set of shrinkage feature points to obtain skeleton point cloud data; Determining the neighborhood point set of each skeleton point in the skeleton point cloud data to obtain a plurality of skeleton point neighborhood sets; Constructing an undirected graph based on each of the skeleton point neighborhood sets, and determining the topological relationship between each skeleton point in each of the skeleton point neighborhood sets according to the undirected graph; Generating the trace lines of the target rock mass according to the topological relationship between each skeleton point in each of the skeleton point neighborhood sets.

[0007] Optionally, the obtaining three-dimensional point cloud data of the target rock mass includes: Collecting structural plane image data of the target rock mass through an image acquisition device; Generating three-dimensional point cloud data of the target rock mass based on the structural plane image data.

[0008] Optionally, the extracting the edge points of the target rock mass from the three-dimensional point cloud data includes: Determine the set of k-nearest neighbor points for each point in the three-dimensional point cloud data; Determine the centroid of the set of k-nearest neighbor points for each point, and determine the resolution of the set of k-nearest neighbor points for each point; For the set of k-nearest neighbor points of each point, determine whether the distance between each point in the set of k-nearest neighbor points and the centroid is greater than a classification threshold; the classification threshold is the product of the resolution of the set of k-nearest neighbor points and a preset parameter; If the distance between a point in the set of k-nearest neighbor points and the centroid is greater than the classification threshold, then take this point as an edge point.

[0009] Optionally, determining the set of neighbor points for each skeleton point in the skeleton point cloud data includes: For each skeleton point, determine the skeleton points in the skeleton point cloud data whose distance from this skeleton point is less than a preset threshold, and obtain the set of neighbor points of this skeleton point.

[0010] Optionally, the nodes in the undirected graph are the skeleton points in the set of skeleton point neighborhoods, and each edge in the undirected graph is attached with a corresponding weight value, and the weight value is the distance between the two skeleton points connected by the corresponding edge.

[0011] Optionally, determining the topological relationship between the skeleton points in each set of skeleton point neighborhoods according to the undirected graph includes: For each set of skeleton point neighborhoods, perform the following processing according to the undirected graph: Set a first point cloud set and a second point cloud set, where the first point cloud set includes a root node R in the undirected graph, and the second point cloud set includes the other nodes in the undirected graph except the root node R; the root node R is the center point of the set of skeleton point neighborhoods; Select a node closest to the root node R from the second point cloud set, denoted as node s, and move the node s from the second point cloud set to the first point cloud set; Calculate the distances from the root node R through the node s to each node in the second point cloud set, and obtain the first distances from the root node R to each node in the second point cloud set; Calculate the distances from the root node R without passing through the node s to each node in the second point cloud set, and obtain the corresponding second distances from the root node R to each node in the second point cloud set; For each node in the second point cloud set, if the second distance from the root node R to this node is less than the first distance, then take the sum of the second distance and the weight value of the edge connecting the root node R and this node in the undirected graph as the new second distance; Loop through the steps of selecting a node s closest to the root node R from the second point cloud set, moving the node s from the second point cloud set to the second point cloud set, calculating the first distance, calculating the second distance, and updating the second distance until all nodes in the second point cloud set are moved to the first point cloud set, obtaining the shortest path from the root node R to each node other than the root node R in the undirected graph, so as to obtain the topological relationship between each skeleton point in each skeleton point neighborhood set.

[0012] Optionally, generating the trace line of the target rock mass according to the topological relationship between each skeleton point in each skeleton point neighborhood set includes: According to the shortest path from the root node R to each node other than the root node R in the undirected graph, connect each node in sequence to form the trace line of the target rock mass.

[0013] In a second aspect, the present invention provides an apparatus for extracting a trace line of a rock mass, including: A data acquisition module, configured to acquire three-dimensional point cloud data of a target rock mass; An edge point extraction module, configured to extract edge points of the target rock mass from the three-dimensional point cloud data to obtain an initial shrinkage feature point set; A skeleton point extraction module, configured to extract L1 median points from the initial shrinkage feature point set by using the L1 median algorithm to obtain skeleton point cloud data; A skeleton point neighborhood determination module, configured to determine a neighborhood point set of each skeleton point in the skeleton point cloud data to obtain a plurality of skeleton point neighborhood sets; A topological relationship determination module, configured to construct an undirected graph according to each skeleton point neighborhood set, and determine the topological relationship between each skeleton point in each skeleton point neighborhood set according to the undirected graph; A trace line generation module, configured to generate the trace line of the target rock mass according to the topological relationship between each skeleton point in each skeleton point neighborhood set.

[0014] In a third aspect, the present invention provides an electronic device, including a memory and a processor; The memory is used to store a computer program; The processor is configured to implement the method for extracting a trace line of a rock mass as described in the first aspect when executing the computer program.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method for extracting a trace line of a rock mass as described in the first aspect is implemented.

[0016] The beneficial effects of a method, device, electronic device and storage medium for extracting rock mass trace lines of the present invention are as follows: Obtain the three-dimensional point cloud data of the target rock mass, and the three-dimensional point cloud data contains the spatial three-dimensional coordinates of the target rock mass. Extract the edge points of the target rock mass in the three-dimensional point cloud data to obtain the initial shrinkage feature point set. Since the rock mass trace line exists at the edge position of the rock mass, extracting the edge points of the target rock mass can reduce the computational amount of subsequent extraction of the rock mass trace line from the point cloud data. Use the L1 median algorithm to extract the L1 median points in the initial shrinkage feature point set to obtain the skeleton point cloud data, and further determine the skeleton points from the extracted edge points. Construct an undirected graph based on the neighborhood sets of each skeleton point, and determine the topological relationship between each skeleton point in the neighborhood sets of each skeleton point according to the undirected graph. The shortest path connecting each skeleton point can be quickly and accurately found through the undirected graph. Generate the trace line of the target rock mass according to the topological relationship between each skeleton point in the neighborhood sets of each skeleton point to efficiently and automatically obtain the accurate trace line of the target rock mass. Moreover, compared with the method of manually measuring the trace line on-site, the present invention accurately extracts the skeleton points from the three-dimensional point cloud data of the target rock mass by using a software algorithm, and combines the undirected graph to determine the topological relationship between each skeleton point, so as to accurately find the shortest path connecting each skeleton point, thereby extracting the accurate trace line of the target rock mass. The whole process does not require manual intervention and does not depend on the experience of the operator, greatly improving the accuracy of rock mass trace line extraction. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flowchart of a method for extracting rock mass trace lines according to an embodiment of the present invention; Figure 2 It is a schematic diagram of extracting skeleton points from the neighborhood in the initial shrinkage feature point set according to an embodiment; Figure 3 It is a schematic diagram of an undirected graph according to an embodiment; Figure 4 It is a flowchart of extracting edge points in the three-dimensional point cloud data according to an embodiment; Figure 5 It is a flowchart of determining the topological relationship between each skeleton point in the neighborhood set of skeleton points according to an embodiment; Figure 6 It is a schematic diagram of the skeleton line connecting each skeleton point according to an embodiment; Figure 7 It is a schematic diagram of the image data of the target rock mass and its corresponding trace line according to an embodiment; Figure 8 It is a schematic structural diagram of a device for extracting rock mass trace lines according to an embodiment of the present invention; Figure 9 It is a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following provides a detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments described herein. Instead, these embodiments are provided to more thoroughly and completely understand the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not used to limit the protection scope of the present invention.

[0019] It should be understood that the various steps described in the method embodiments of the present invention can be executed in different orders and / or executed in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this regard.

[0020] As used herein, the term "including" and its variants are open-ended, that is, "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts such as "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules, or units, and are not used to limit the order of functions executed by these devices, modules, or units or their interdependent relationships.

[0021] It should be noted that the modification of "one" and "multiple" mentioned in the present invention is illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly specified in the context, it should be understood as "one or more".

[0022] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0023] As Figure 1 shown, a method for extracting rock mass traces provided by an embodiment of the present invention includes: S100: Obtain the three-dimensional point cloud data of the target rock mass.

[0024] In some embodiments, the structural plane image data of the target rock mass is collected by an image acquisition device; based on the structural plane image data, the three-dimensional point cloud data of the target rock mass is generated. Among them, the image acquisition device can be an unmanned aerial vehicle camera device or a high-definition camera.

[0025] Specifically, high-resolution three-dimensional point cloud data is generated from the structural plane image data through the SIFT algorithm.

[0026] Specifically, the structural plane image data is a photo of the structural plane, and the structural plane can be the slope of the target rock mass.

[0027] Specifically, the three-dimensional point cloud data is three-dimensional coordinate data in space.

[0028] S200: Extract the edge points of the target rock mass from the three-dimensional point cloud data to obtain the initial set of contraction feature points.

[0029] Specifically, through analysis, it is found that the rock mass trace is often located at the edge of the rock mass. Therefore, edge points can be extracted from the three-dimensional point cloud data first, and then subsequent processing can be carried out based on the extracted edge points. The three-dimensional point cloud data contains the point clouds corresponding to various structures of the target rock mass, making the data volume of the three-dimensional point cloud data too large. Directly extracting the skeleton points corresponding to the trace from the three-dimensional point cloud data will result in too much computational effort, thus affecting the computational efficiency. Therefore, in this embodiment, edge points are extracted from the three-dimensional point cloud data first to reduce the data computational effort for subsequent extraction of skeleton points.

[0030] S300: Use the L1 median algorithm to extract the L1 median points from the initial set of contraction feature points to obtain the skeleton point cloud data.

[0031] Specifically, the expression of the L1 median algorithm is as follows: ; ; In the formula, represents the Gaussian weight function, E represents the initial set of contraction feature points, represents the initial set of contraction feature points E in one point, , m is the initial set of contraction feature points E the number of points in, represents the source input point in the initial set of contraction feature points E , is the point to the source input point distance, represents the parameter controlling the neighborhood size, argmin() represents finding the parameter value that makes the objective function reach the minimum value, X represents the L1 median point. The L1 median point is a skeleton point in the skeleton point cloud data.

[0032] Such as Figure 2As shown, the L1 median algorithm is used to extract the L1 median points from the initial shrunk feature point set, that is, the initial shrunk feature point set is divided into multiple neighborhoods B, and the L1 median points are extracted in each neighborhood B, and the L1 median point is the skeleton point B1.

[0033] S400: Determine the neighborhood point set of each skeleton point in the skeleton point cloud data to obtain multiple skeleton point neighborhood sets.

[0034] Specifically, after obtaining the skeleton point cloud data, the skeleton point cloud data is saved, then the neighborhoods of all points in the skeleton point cloud data are calculated, and the calculated neighborhoods are put into a new set to obtain multiple skeleton point neighborhood sets.

[0035] S500: Construct an undirected graph according to each skeleton point neighborhood set, and determine the topological relationship between each skeleton point in each skeleton point neighborhood set according to the undirected graph.

[0036] Specifically, as Figure 3 shown, the undirected graph consists of a set of nodes (A1, A2, A3, A4, A5, A6) and a set of edges. Among them, the edges have no directionality. The nodes in the undirected graph are the skeleton points in the skeleton point neighborhood set, and each edge in the undirected graph is attached with a corresponding weight value, and the weight value is the distance between the two skeleton points connected by the corresponding edge.

[0037] S600: Generate the trace line of the target rock mass according to the topological relationship between each skeleton point in each skeleton point neighborhood set.

[0038] In some embodiments, according to the determined topological relationship between each skeleton point in each skeleton point neighborhood set, connecting each skeleton point in sequence can obtain the trace line of the target rock mass.

[0039] In this embodiment, the three-dimensional point cloud data of the target rock mass is obtained, and the three-dimensional point cloud data contains the spatial three-dimensional coordinates of the target rock mass. The edge points of the target rock mass in the three-dimensional point cloud data are extracted to obtain the initial shrunk feature point set. Since the trace line of the rock mass exists at the edge position of the rock mass, extracting the edge points of the target rock mass can reduce the calculation amount of extracting the trace line of the rock mass from the point cloud data subsequently. The L1 median algorithm is used to extract the L1 median points from the initial shrunk feature point set to obtain the skeleton point cloud data, and further determine the skeleton points from the extracted edge points. An undirected graph is constructed according to each skeleton point neighborhood set, and according to the undirected graph, the topological relationship between each skeleton point in each skeleton point neighborhood set is determined. The shortest path connecting each skeleton point can be quickly and accurately found through the undirected graph. The trace line of the target rock mass is generated according to the topological relationship between each skeleton point in each skeleton point neighborhood set to efficiently and automatically obtain the accurate trace line of the target rock mass.

[0040] Optionally, asFigure 4 As shown in Figure 4 , the steps of extracting the edge points of the target rock mass from the 3D point cloud data in S200 to obtain the initial shrinkage feature point set include the following steps: S410: Determine the k-neighborhood point set of each point in the 3D point cloud data.

[0041] S420: Determine the centroid of the k-neighborhood point set of each point, and determine the resolution of the k-neighborhood point set of each point.

[0042] S430: For the k-neighborhood point set of each point, judge whether the distance between each point in the k-neighborhood point set and the centroid is greater than the classification threshold; wherein, the classification threshold is the product of the resolution of the k-neighborhood point set and the preset parameter.

[0043] S440: If the distance between a point in the k-neighborhood point set and the centroid is greater than the classification threshold, then take this point as an edge point.

[0044] Specifically, the K-D tree method is used to calculate the k-neighborhood point set of each point in the 3D point cloud data. For the k-neighborhood point set of any point , randomly select a point as the centroid point of the k-neighborhood point set , and calculate the centroid of the k-neighborhood point set by taking the mean of adjacent points . Determine the centroid according to the following expression : ; In the formula, k represents the number of neighborhood points in the k-neighborhood point set , represents a k-neighborhood point set, represents the k-neighborhood point set , represents the coordinates of any point in the k-neighborhood point set , j represents the coordinates of the centroid of the k-neighborhood point set,

[0045] Specifically, in order to adapt to the change of density, calculate the resolution of the k-neighborhood point set according to the following expression : ; In the formula, , represent any two points in the k-neighborhood point set .

[0046] Since the local density of the point is considered separately for each point, the evaluation of ensures the scale invariance.

[0047] Specifically, multiply by a fixed preset parameter λ as the classification threshold. If the distance between the point and the centroid is greater than the classification threshold, then the point is classified as an edge point, and the expression is as follows: ; where λ is the preset parameter, is the resolution of the k-neighborhood point set , represents any point in the k-neighborhood point set , represents the centroid of the k-neighborhood point set .

[0048] In this optional embodiment, the k-neighborhood point set of each point in the three-dimensional point cloud data is calculated, and then the centroid of each k-neighborhood point set is determined. Based on the centroid of the k-neighborhood point set, the resolution of the k-neighborhood point set is calculated to adapt to the change of density; the resolution of the k-neighborhood point set is used as the parameter of the classification threshold for edge point classification. Since the local density of each point in the three-dimensional point cloud data is considered separately, the edge point classification based on the resolution ensures the scale invariance.

[0049] Optionally, in S400, determining the neighborhood point set of each skeleton point in the skeleton point cloud data includes: For each skeleton point, determine the skeleton points in the skeleton point cloud data whose distance from the skeleton point is less than the preset threshold to obtain the neighborhood point set of the skeleton point.

[0050] Specifically, the K-D tree method is used to determine the neighborhood corresponding to each skeleton point in the skeleton point cloud data. Specifically, a preset threshold can be set. For each skeleton point, calculate the distance between the skeleton point and other skeleton points in the skeleton point cloud data, and form a neighborhood point set of the skeleton point with the skeleton points whose distance is less than the preset threshold.

[0051] In this optional embodiment, determining the neighborhood of each skeleton point in the skeleton point cloud data provides data support for constructing the undirected graph corresponding to the neighborhood point set of each skeleton point subsequently.

[0052] Optionally, as Figure 5 shown, in S500, an undirected graph is constructed according to each skeleton point neighborhood set, and according to the undirected graph, the topological relationship between each skeleton point in each skeleton point neighborhood set is determined, including the following steps: For each skeleton point neighborhood set, perform the following processing according to the undirected graph: S510: Set a first point cloud set and a second point cloud set, where the first point cloud set includes a root node R in an undirected graph, and the second point cloud set includes other nodes in the undirected graph except the root node R. Among them, the root node R is the center point of the skeleton point neighborhood set.

[0053] S520: Select a node closest to the root node R from the second point cloud set, denoted as node s, and move node s from the second point cloud set to the first point cloud set.

[0054] S530: Calculate the distances from the root node R through node s to each node in the second point cloud set to obtain the first distances from the root node R to each node in the second point cloud set.

[0055] S540: Calculate the distances from the root node R without passing through node s to each node in the second point cloud set to obtain the second distances corresponding to each node in the second point cloud set from the root node R.

[0056] S550: For each node in the second point cloud set, if the second distance from the root node R to the node is less than the first distance, then use the sum of the second distance and the weight value of the edge connecting node R and the node in the undirected graph as the new second distance.

[0057] S560: Iteratively loop through steps S520 to S550 until all nodes in the second point cloud set are moved to the first point cloud set. Connect each node except the root node R in the undirected graph in the order of the points in the second point cloud set moving to the first point cloud set starting from the root node R, and the shortest path from the root node R to each node except node R in the undirected graph is obtained.

[0058] Integrate the shortest paths from the root node R to each node except node R in the undirected graph corresponding to each skeleton point neighborhood set, and the shortest paths of all skeleton points are obtained. Connect each skeleton point according to the shortest paths of all skeleton points, and the corresponding trace line is obtained, as Figure 6 shown.

[0059] In this optional embodiment, by circularly calculating the path distances from the root node R to each skeleton point, a shortest path from the root node R to each skeleton point is formed. This shortest path is the topological relationship between each skeleton point, enabling the efficient determination of the topological relationship between each skeleton point in the skeleton point neighborhood set.

[0060] As Figure 7 shown, Figure 7 are the image data of the target rock mass and its corresponding trace line.

[0061] In this alternative embodiment, by determining the shortest paths from the root node R to each skeleton point, the trace of the target rock mass can be quickly and accurately obtained.

[0062] As Figure 8 shown, an extraction device 800 for the trace of a rock mass provided by an embodiment of the present invention includes: A data acquisition module 810, configured to acquire three-dimensional point cloud data of a target rock mass.

[0063] An edge point extraction module 820, configured to extract edge points of the target rock mass from the three-dimensional point cloud data to obtain an initial set of contraction feature points.

[0064] A skeleton point extraction module 830, configured to extract L1 median points from the initial set of contraction feature points by using the L1 median algorithm to obtain skeleton point cloud data.

[0065] A skeleton point neighborhood determination module 840, configured to determine a neighborhood point set of each skeleton point in the skeleton point cloud data to obtain a plurality of skeleton point neighborhood sets.

[0066] A topological relationship determination module 850, configured to construct an undirected graph according to each skeleton point neighborhood set, and determine the topological relationship between each skeleton point in each skeleton point neighborhood set according to the undirected graph.

[0067] A trace generation module 860, configured to generate a trace of the target rock mass according to the topological relationship between each skeleton point in each skeleton point neighborhood set.

[0068] Optionally, the acquiring of the three-dimensional point cloud data of the target rock mass includes: Collecting structural plane image data of the target rock mass by an image acquisition device; Generating three-dimensional point cloud data of the target rock mass based on the structural plane image data.

[0069] Optionally, the extracting of the edge points of the target rock mass from the three-dimensional point cloud data includes: Determining a k-neighborhood point set of each point in the three-dimensional point cloud data; Determining the centroid of the k-neighborhood point set of each point, and determining the resolution of the k-neighborhood point set of each point; For the k-neighborhood point set of each point, determining whether the distance between each point in the k-neighborhood point set and the centroid is greater than a classification threshold; the classification threshold is the product of the resolution of the k-neighborhood point set and a preset parameter; If the distance between a point in the k-neighborhood point set and the centroid is greater than the classification threshold, then taking this point as an edge point.

[0070] Optionally, determining the set of neighborhood points of each skeleton point in the skeleton point cloud data includes: For each skeleton point, determine the skeleton points in the skeleton point cloud data whose distance from this skeleton point is less than a preset threshold, to obtain the set of neighborhood points of this skeleton point.

[0071] Optionally, the nodes in the undirected graph are the skeleton points in the skeleton point neighborhood set, and each edge of the undirected graph is attached with a corresponding weight value, and the weight value is the distance between the two skeleton points connected by the corresponding edge.

[0072] Optionally, according to the undirected graph, determining the topological relationship between each skeleton point in each skeleton point neighborhood set includes: For each skeleton point neighborhood set, perform the following processing according to the undirected graph: Set a first point cloud set and a second point cloud set, where the first point cloud set includes a root node R in the undirected graph, and the second point cloud set includes other nodes in the undirected graph except the root node R; Select a node closest to the root node R from the second point cloud set, denoted as node s, and move the node s from the first point cloud set to the second point cloud set; Calculate the distances from the root node R through the node s to each node in the second point cloud set, to obtain the first distances from the root node R to each node in the second point cloud set; Calculate the distances from the root node R without passing through the node s to each node in the second point cloud set, to obtain the second distances corresponding to each node in the second point cloud set from the root node R; For each node in the second point cloud set, if the second distance from the root node R to this node is less than the first distance, then use the sum of the second distance and the weight value of the edge connecting the root node R and this node in the undirected graph as the new second distance; Loop through the steps of selecting a node s closest to the root node R from the second point cloud set, moving the node s from the second point cloud set to the second point cloud set, calculating the first distance, calculating the second distance, and updating the second distance, until all nodes in the first point cloud set are moved to the second point cloud set, to obtain the shortest paths from the root node R to each node other than the root node R in the undirected graph, so as to obtain the topological relationship between each skeleton point in each skeleton point neighborhood set.

[0073] Optionally, generating the trace line of the target rock mass according to the topological relationship between each skeleton point in each skeleton point neighborhood set includes: According to the shortest paths from the root node R to each node in the undirected graph except the root node R, connect each node in sequence to form the trace of the target rock mass.

[0074] As Figure 9 shown, an electronic device 900 provided by an embodiment of the present invention includes a memory 910 and a processor 920; the memory 910 is used to store a computer program; the processor 920 is used to implement the method for extracting the trace of the rock mass as described above when executing the computer program.

[0075] Or rather, an electronic device 900 includes a memory 910 and a processor 920 coupled to the memory 910; the memory 910 is configured to store a computer program; the processor 920 is configured to perform the following operations when executing the computer program: Obtain the three-dimensional point cloud data of the target rock mass; Extract the edge points of the target rock mass in the three-dimensional point cloud data to obtain an initial set of shrinkage feature points; Use the L1 median algorithm to extract the L1 median points in the initial set of shrinkage feature points to obtain the skeleton point cloud data; Determine the set of neighborhood points of each skeleton point in the skeleton point cloud data to obtain a plurality of skeleton point neighborhood sets; Construct an undirected graph according to each of the skeleton point neighborhood sets, and determine the topological relationship between each skeleton point in each of the skeleton point neighborhood sets according to the undirected graph; Generate the trace of the target rock mass according to the topological relationship between each skeleton point in each of the skeleton point neighborhood sets.

[0076] A computer-readable storage medium provided by an embodiment of the present invention has a computer program stored thereon, and when the computer program is executed by a processor, the method for extracting the trace of the rock mass as described above is implemented.

[0077] Or rather, a non-volatile computer-readable storage medium has a computer program stored thereon, and when the computer program is executed by a processor, the processor is caused to perform the following operations: Obtain the three-dimensional point cloud data of the target rock mass; Extract the edge points of the target rock mass in the three-dimensional point cloud data to obtain an initial set of shrinkage feature points; Use the L1 median algorithm to extract the L1 median points in the initial set of shrinkage feature points to obtain the skeleton point cloud data; Determine the set of neighborhood points of each skeleton point in the skeleton point cloud data to obtain a plurality of skeleton point neighborhood sets; Construct an undirected graph based on each set of skeleton point neighborhoods, and determine the topological relationships between the skeleton points in each set of skeleton point neighborhoods according to the undirected graph; Generate the trace line of the target rock mass according to the topological relationships between the skeleton points in each set of skeleton point neighborhoods.

[0078] Now, an electronic device 900 that can be a server or a client of the present invention will be described. It is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device 900 is intended to represent various forms of digital electronic computer devices, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device 900 can also represent various forms of mobile devices, such as, personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0079] The electronic device 900 includes a computing unit that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for device operation can also be stored. The computing unit, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0080] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc. In the present application, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiments of the present invention. In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0081] Although the present invention is disclosed as above, the scope of protection of the present invention is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will all fall within the scope of protection of the present invention.

Claims

1. A method for extracting rock mass traces, characterized in that: include: Obtain three-dimensional point cloud data of the target rock mass; Extracting edge points of the target rock mass in the three-dimensional point cloud data to obtain an initial shrinkage feature point set; Using an L1 median algorithm to extract L1 median points from the initial contraction feature point set to obtain skeleton point cloud data; Determine a neighborhood point set of each skeleton point in the skeleton point cloud data to obtain a plurality of skeleton point neighborhood sets; Constructing an undirected graph according to each of the skeleton point neighborhood sets, and determining the topological relationship between each of the skeleton points in each of the skeleton point neighborhood sets according to the undirected graph; The trace of the target rock mass is generated according to the topological relationship between each skeleton point in the neighborhood set of each skeleton point.

2. The method for extracting rock mass traces according to claim 1, characterized in that: The acquisition of three-dimensional point cloud data of the target rock mass comprises: Collecting structural surface image data of the target rock mass by an image acquisition device; Based on the structural surface image data, three-dimensional point cloud data of the target rock mass is generated.

3. The method for extracting rock mass traces according to claim 1, characterized in that: The extracting edge points of the target rock mass in the three-dimensional point cloud data comprises: Determine a k-neighborhood point set for each point in the three-dimensional point cloud data; Determine the centroid of the k-neighborhood point set of each point, and determine the resolution of the k-neighborhood point set of each point; For each point's k-neighborhood point set, determine whether the distance between each point in the k-neighborhood point set and the centroid is greater than a classification threshold; the classification threshold is the product of the resolution of the k-neighborhood point set and a preset parameter; If the distance between a point in the k-neighborhood point set and the centroid is greater than a classification threshold, the point is regarded as an edge point.

4. The method for extracting rock mass traces according to claim 1, characterized in that: Determining a neighborhood point set for each skeleton point in the skeleton point cloud data comprises: For each skeleton point, skeleton points whose distances to the skeleton point are less than a preset threshold are determined in the skeleton point cloud data to obtain a neighborhood point set of the skeleton point.

5. The rock mass trace extraction method according to claim 1, characterized in that: The nodes in the undirected graph are skeleton points in the skeleton point neighborhood set. Each edge of the undirected graph is attached with a corresponding weight, and the weight is the distance between two skeleton points connected by the corresponding edge.

6. The method for extracting rock mass traces according to claim 5, characterized in that: Determining the topological relationship between each skeleton point in each skeleton point neighborhood set according to the undirected graph includes: For each of the skeleton point neighborhood sets, the following processing is performed according to the undirected graph: Setting a first point cloud set and a second point cloud set, wherein the first point cloud set includes a root node R in the undirected graph, and the second point cloud set includes other nodes in the undirected graph except the root node R; the root node R is the center point of the skeleton point neighborhood set; Select a node closest to the root node R from the second point cloud set, denoted as node s, and move the node s from the second point cloud set to the first point cloud set; Calculate the distance from the root node R through the node s to each node in the second point cloud set to obtain a first distance from the root node R to each node in the second point cloud set; Calculate the distance from the root node R to each node in the second point cloud set without passing through the node s, and obtain the second distance corresponding to the root node R to each node in the second point cloud set; For each node in the second point cloud set, if the second distance from the root node R to the node is less than the first distance, the sum of the second distance and the weight corresponding to the edge connecting the root node R and the node in the undirected graph is taken as the new second distance; The steps of selecting a node s closest to node R from the second point cloud set, moving the node s from the second point cloud set to the first point cloud set, calculating the first distance, calculating the second distance and updating the second distance are performed in a loop until all nodes in the second point cloud set are moved to the first point cloud set, and the shortest path from the root node R to each node in the undirected graph except the root node R is obtained to obtain the topological relationship between each skeleton point in the neighborhood set of each skeleton point.

7. The method for extracting rock mass traces according to claim 6, characterized in that: Generating the trace of the target rock mass according to the topological relationship between each skeleton point in the neighborhood set of each skeleton point comprises: According to the shortest path from the root node R to each node except the root node R in the undirected graph, each node is connected in sequence to form the trace of the target rock mass.

8. A rock mass trace extraction device, characterized in that: include: A data acquisition module, used to acquire three-dimensional point cloud data of the target rock mass; An edge point extraction module is used to extract edge points of the target rock mass in the three-dimensional point cloud data to obtain an initial shrinkage feature point set; A skeleton point extraction module, used to extract L1 median points from the initial contraction feature point set using an L1 median algorithm to obtain skeleton point cloud data; A skeleton point neighborhood determination module, used to determine a neighborhood point set of each skeleton point in the skeleton point cloud data to obtain a plurality of skeleton point neighborhood sets; A topological relationship determination module, used to construct an undirected graph according to each of the skeleton point neighborhood sets, and determine the topological relationship between each of the skeleton points in each of the skeleton point neighborhood sets according to the undirected graph; The trace generation module is used to generate the trace of the target rock mass according to the topological relationship between each skeleton point in the neighborhood set of each skeleton point.

9. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to implement the rock mass trace extraction method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method for extracting rock mass traces as described in any one of claims 1 to 7 is implemented.

Images