Mobile measurement point cloud surface reconstruction method and system based on tunnel scene

Abstract

The invention relates to a mobile measurement point cloud surface reconstruction method and system based on a tunnel scene. The method comprises the following steps: acquiring three-dimensional original point cloud data in the tunnel scene by adopting mobile measurement equipment; converting the three-dimensional original point cloud data into two-dimensional plane point cloud data through expansion projection; performing Delaunay triangulation on the two-dimensional plane point cloud data to construct a two-dimensional triangulation network; according to vertex association, the vertex of each two-dimensional triangulation network is marked at a unique corresponding point in the three-dimensional point cloud data; and mapping the topological relation of the two-dimensional triangulation network into the three-dimensional original point cloud, directly converting the two-dimensional adjacent edges into three-dimensional adjacent edges through an edge mapping rule, and generating a tunnel three-dimensional surface model. According to the method, the three-dimensional original point cloud is expanded, projected and converted into the two-dimensional plane point cloud, and then Delaunay triangularization is performed, so that the time required by tunnel point cloud network construction is greatly shortened; and the generation probability of isolated triangulation networks and non-closed holes is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional point cloud surface reconstruction, and in particular to a method and system for reconstructing a point cloud surface using mobile measurement based on a tunnel scene. Background Art

[0002] In the field of 3D point cloud surface reconstruction, existing surface reconstruction methods include traditional methods and deep learning methods. Traditional methods include explicit surface reconstruction and implicit surface reconstruction. Deep learning methods include end-to-end neural networks, implicit neural representations, and hybrid network methods.

[0003] When faced with large-scale point clouds in tunnel scenes, existing surface reconstruction methods have low network efficiency (the time complexity of 3D network construction is about O( ), and the environment inside the tunnel is relatively complex, often with evacuation platforms, cable supports and other ancillary equipment. The laser radar's scanning line is easily blocked and intercepted by these devices, resulting in a sudden change in the distance between adjacent laser points, forming a "fault zone" in the 3D point cloud. As a result, when the original 3D point cloud is directly meshed, it is easy to form scattered isolated triangulated networks. Moreover, due to the insufficient point density inside the mutation area, the meshing algorithm is difficult to fill the effective topology, resulting in non-closed holes.

[0004] Therefore, it is necessary to provide a new method for surface reconstruction of mobile measurement point clouds based on tunnel scenes. Summary of the Invention

[0005] Based on the above problems existing in the prior art, the purpose of the embodiments of the present invention is to provide a method and system for surface reconstruction of mobile measurement point cloud based on tunnel scenes. By converting the original three-dimensional point cloud into a two-dimensional plane point cloud through unfolding projection, and then performing Delaunay triangulation, the network construction calculation complexity is reduced from O( ) is reduced to a two-dimensional network , which greatly reduces the time required for tunnel point cloud construction and reduces the probability of generating isolated triangulated networks and non-closed holes.

[0006] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is: a method for surface reconstruction of mobile measurement point cloud based on tunnel scene, comprising: S1, using mobile measurement equipment to obtain 3D raw point cloud data in the tunnel scene; S2, converting the 3D original point cloud data into 2D plane point cloud data through unfolding projection; S3, performing Delaunay triangulation on the two-dimensional plane point cloud data to construct a two-dimensional triangulated network; S4, based on vertex association, marking each vertex of the 2D triangulated network at a unique corresponding point in the 3D point cloud data; S5, maps the topological relationship of the 2D triangulated network to the 3D original point cloud, and directly converts the 2D adjacent edges into 3D adjacent edges through the edge mapping rule to generate the 3D surface model of the tunnel.

[0007] Furthermore, the three-dimensional original point cloud data of the tunnel scene is obtained using a mobile measuring device equipped with a laser radar. The laser radar of the mobile measuring device scans along the longitudinal axis of the tunnel, and the cross-section is orthogonal to the longitudinal axis of the tunnel, so that the three-dimensional original point cloud distribution has axial continuity. According to the laser pulse emitted by the laser radar, the time difference of the reflected light is measured to determine the distance, thereby obtaining a large amount of three-dimensional original point cloud data.

[0008] Furthermore, the converting of the three-dimensional original point cloud data into two-dimensional plane point cloud data by unfolding and projecting includes: Step S21, converting the three-dimensional original point cloud data into two-dimensional polar coordinates; Step S22 , expanding the points in the two-dimensional polar coordinates into the two-dimensional plane, so as to map the three-dimensional points of the three-dimensional original point cloud data into the two-dimensional plane coordinates to obtain the two-dimensional plane point cloud data.

[0009] Furthermore, the three-dimensional original point cloud data is converted into two-dimensional polar coordinates, including: The three-dimensional original point cloud data uses the tunnel longitudinal axis as the reference line to establish the global coordinate system as follows: x,y,z ),in, z The axis corresponds to the longitudinal axis of the tunnel, and the cross section is located at xy On a plane; Project each point in the 3D original point cloud data into the polar coordinate system ( ), is the rotation angle, is the radial distance; Radial distance The calculation formula is:

[0010] in ， is the radial distance, x is the horizontal coordinate of the original 3D point cloud data, y is the vertical coordinate of the original 3D point cloud data; Rotation angle The calculation formula is:

[0011] in ， is the rotation angle, x is the horizontal coordinate of the original 3D point cloud data, y is the vertical coordinate of the original 3D point cloud data; The method of expanding the points in the two-dimensional polar coordinates into the two-dimensional plane to map the three-dimensional points of the three-dimensional original point cloud data into the two-dimensional plane coordinates to obtain the two-dimensional plane point cloud data includes: The rotation angle in two-dimensional polar coordinates Corresponding to the horizontal axis of the plane, the radial distance in the two-dimensional polar coordinates The horizontal axis and vertical axis of the corresponding plane, for the coordinates of point P in the three-dimensional original point cloud data, are ( x,y,z ), the corresponding two-dimensional polar coordinate expansion point The coordinates of , ),; Project each point in the 3D original point cloud data into the 2D polar coordinate system to obtain the corresponding 2D coordinate point ( , ), the points in the polar coordinate system are expanded to a two-dimensional plane to form two-dimensional plane point cloud data, so that the three-dimensional original point cloud data distributed in the three-dimensional space is converted into two-dimensional plane point cloud data in the two-dimensional space.

[0012] Furthermore, the Delaunay triangulation of the two-dimensional plane point cloud data to construct a two-dimensional triangulated network includes: Step S31, preprocessing the two-dimensional plane point cloud data, including removing noise points and abnormal points, to improve the quality and efficiency of triangulation; Step S32, selecting three non-collinear point coordinates in the two-dimensional plane point cloud data to construct an initial triangle; Step S33, using the divide-and-conquer algorithm to process large-scale two-dimensional point clouds, the time complexity is , which is superior to the point-by-point interpolation method, performs local density compensation on the point cloud in the fault zone area, generates virtual points using linear interpolation according to the density of neighboring points, and gradually interpolates the remaining point cloud points according to the empty circle property criterion of Delaunay triangulation.

[0013] Furthermore, the method of marking each vertex of the two-dimensional triangulated network at a unique corresponding point in the three-dimensional point cloud data according to vertex association includes: when the three-dimensional original point cloud data is converted into two-dimensional plane point cloud data through unfolding projection, each point in the two-dimensional plane point cloud data has a corresponding two-dimensional coordinate ( ) and the three-dimensional coordinate points of the three-dimensional original point cloud data ( x,y,z After the 2D triangulation is completed, a one-to-one correspondence is established between the 2D triangulation vertices and the 3D coordinate points of the 3D original point cloud data, and each vertex of the 2D triangulation is marked with a unique 3D coordinate point in the 3D original point cloud data ( x,y,z ).

[0014] Furthermore, the topological relationship of the two-dimensional triangulated network is mapped into the three-dimensional original point cloud, and the two-dimensional adjacent edges are directly converted into three-dimensional adjacent edges through the edge mapping rule to generate the three-dimensional surface model of the tunnel, including: Step S51: converting adjacent edges of the two-dimensional triangulated mesh into three-dimensional adjacent edges.

[0015] Step S52: combine all three-dimensional adjacent edges to generate a closed triangular mesh surface, and then generate a three-dimensional surface model of the tunnel.

[0016] A mobile measurement point cloud surface reconstruction system based on a tunnel scene is applied to the above-mentioned mobile measurement point cloud surface reconstruction method based on a tunnel scene, and the system includes: A three-dimensional data acquisition module is used to acquire three-dimensional original point cloud data in the tunnel scene using a mobile measurement device; The data projection conversion module is used to convert the three-dimensional original point cloud data into two-dimensional plane point cloud data through unfolding projection; 2D triangulation module, used to perform Delaunay triangulation on 2D plane point cloud data to construct 2D triangulation network; A vertex association module is used to mark each vertex of the two-dimensional triangulation network with a unique corresponding point in the three-dimensional point cloud data based on vertex association; The mapping output module is used to map the topological relationship of the 2D triangulated network to the 3D original point cloud, and directly convert the 2D adjacent edges into 3D adjacent edges through the edge mapping rule to generate the 3D surface model of the tunnel. The embodiment of the present invention further provides a network-side server, comprising: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the above-mentioned mobile measurement point cloud surface reconstruction method based on the tunnel scene.

[0017] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned method for surface reconstruction of mobile measurement point cloud based on a tunnel scene.

[0018] The beneficial effects of the present invention are as follows: the present invention provides a method for reconstructing the surface of a mobile measurement point cloud based on a tunnel scene, comprising: using a mobile measurement device to obtain three-dimensional original point cloud data in the tunnel scene; converting the three-dimensional original point cloud data into two-dimensional plane point cloud data through unfolding and projection; performing Delaunay triangulation on the two-dimensional plane point cloud data to construct a two-dimensional triangulated network; marking each vertex of the two-dimensional triangulated network with a unique corresponding point in the three-dimensional point cloud data according to vertex association; mapping the topological relationship of the two-dimensional triangulated network to the three-dimensional original point cloud, and directly converting the two-dimensional adjacent edges into three-dimensional adjacent edges through edge mapping rules to generate a three-dimensional surface model of the tunnel. The present invention provides a method for reconstructing the surface of a mobile measurement point cloud based on a tunnel scene, by converting the three-dimensional original point cloud into a two-dimensional plane point cloud through unfolding and projection, and then performing Delaunay triangulation, thereby reducing the computational complexity of the network construction from O( ) is reduced to a two-dimensional network , greatly reducing the time required for tunnel point cloud construction; utilizing the fully connected characteristics of the two-dimensional plane, the proportion of isolated triangulated networks and non-closed holes is close to zero, effectively filling the fracture zone area caused by occlusion, and avoiding isolated triangulated networks and non-closed holes; in view of the geometric closure and scanning center fixity of the tunnel scene, a point cloud geometry reconstruction method based on cross-dimensional topological relationship transmission is proposed, which conforms to the characteristics of tunnel point cloud data and has high practicality and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The present invention will be further described below with reference to the accompanying drawings and examples.

[0020] In the picture: Figure 1 Flowchart of the surface reconstruction method of mobile measurement point cloud based on tunnel scene provided by the first embodiment of the present invention; Figure 2 A schematic diagram of a 3D original point cloud of a subway tunnel provided by the first embodiment of the present invention; Figure 3 A schematic diagram of a subway tunnel 3D point cloud projected onto a 2D plane provided by the first embodiment of the present invention; Figure 4 Schematic diagram of the results of local Delaunay triangulation of a two-dimensional point cloud and three-dimensional surface reconstruction provided by the first embodiment of the present invention; Figure 5 A schematic diagram of modules of a mobile measurement point cloud surface reconstruction system based on a tunnel scenario provided by a second embodiment of the present invention; Figure 6 It is a structural diagram of a network-side server provided according to a third embodiment of the present invention. DETAILED DESCRIPTION

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0022] First embodiment: The first embodiment of the present invention provides a method for reconstructing the surface of a mobile measurement point cloud based on a tunnel scene, comprising: using a mobile measurement device to obtain three-dimensional original point cloud data in the tunnel scene; converting the three-dimensional original point cloud data into two-dimensional plane point cloud data through unfolding and projection; performing Delaunay triangulation on the two-dimensional plane point cloud data to construct a two-dimensional triangulated network; marking each vertex of the two-dimensional triangulated network with a unique corresponding point in the three-dimensional point cloud data according to vertex association; mapping the topological relationship of the two-dimensional triangulated network to the three-dimensional original point cloud, and directly converting the two-dimensional adjacent edges into three-dimensional adjacent edges through edge mapping rules to generate a three-dimensional surface model of the tunnel. The method for reconstructing the surface of a mobile measurement point cloud based on a tunnel scene of the present invention converts the three-dimensional original point cloud into a two-dimensional plane point cloud through unfolding and projection, and then performs Delaunay triangulation, thereby reducing the computational complexity of the network construction from about O( ) is reduced to a two-dimensional network , greatly reducing the time required for tunnel point cloud construction; utilizing the fully connected characteristics of the two-dimensional plane, the proportion of isolated triangulated networks and non-closed holes is close to zero, effectively filling the fracture zone area caused by occlusion, and avoiding isolated triangulated networks and non-closed holes; in view of the geometric closure and scanning center fixity of the tunnel scene, a point cloud geometry reconstruction method based on cross-dimensional topological relationship transmission is proposed, which conforms to the characteristics of tunnel point cloud data and has high practicality and adaptability.

[0023] The following is a detailed description of the implementation details of the mobile measurement point cloud surface reconstruction method based on the tunnel scene of this embodiment. The following content is only for the convenience of understanding the implementation details provided, and is not necessary for the implementation of this solution. The specific process of this embodiment is as follows Figure 1 shown.

[0024] Step S1: Using a mobile measurement device to obtain three-dimensional original point cloud data in a tunnel scene.

[0025] Specifically, if Figure 2As shown, the three-dimensional original point cloud data of the tunnel scene is usually obtained by a mobile measuring device equipped with a laser radar. The laser radar of the mobile measuring device scans along the longitudinal axis of the tunnel (i.e., the direction of tunnel extension or track extension), and the cross section (i.e., the plane perpendicular to the longitudinal axis) is orthogonal to the longitudinal axis of the tunnel, thereby making the three-dimensional original point cloud distribution have axial continuity. According to the laser pulse emitted by the laser radar, the time difference of the reflected light is measured to determine the distance, thereby obtaining a large amount of three-dimensional original point cloud data.

[0026] Furthermore, when acquiring the three-dimensional original point cloud data of the tunnel scene, the mobile measurement equipment needs to move at a uniform speed along the longitudinal axis of the tunnel to ensure that the scanning cross-section is strictly orthogonal to the longitudinal axis. The scanning frequency of the lidar should be dynamically adjusted according to the tunnel size and point cloud density requirements to avoid data redundancy or insufficiency.

[0027] However, due to the complexity of tunnel scenes, for example, the presence of ancillary equipment such as evacuation platforms and cable brackets in the tunnel, the scanning line of the lidar is easily blocked and interrupted, resulting in a sudden change in the distance between adjacent laser points, forming a fracture zone in the three-dimensional original point cloud.

[0028] Step S2: Convert the three-dimensional original point cloud data into two-dimensional plane point cloud data through unfolding projection.

[0029] Specifically, the specific steps of converting the three-dimensional original point cloud data into two-dimensional plane point cloud data through unfolding projection include: Step S21 : converting the three-dimensional original point cloud data into two-dimensional polar coordinates.

[0030] Specifically, because tunnel scenes are geometrically closed and have a fixed scanning center, the cross-section of the tunnel scene is a strictly closed curve, such as a circle, horseshoe, or quasi-rectangle. The longitudinal axis of the tunnel scene is approximately a straight line over a short distance, and the point cloud distribution along the longitudinal axis is periodic. When the mobile measurement device's LiDAR scans, the point cloud of each cross-section is distributed in a circular pattern in three-dimensional space. The geometric closure and fixed scanning center ensure axial consistency in the point cloud distribution. The mobile measurement device's LiDAR scan cross-section is orthogonal to the tunnel's longitudinal axis, thereby expanding the original three-dimensional point cloud along the tunnel's longitudinal axis.

[0031] The three-dimensional original point cloud data uses the tunnel longitudinal axis as the reference line to establish the global coordinate system as follows: x,y,z ),in, z The axis corresponds to the longitudinal axis of the tunnel, and the cross section is located at xy On the plane.

[0032] Project each point in the 3D original point cloud data into the polar coordinate system ( ), among which, is the rotation angle, is the radial distance. That is, the radial distance represents the distance from the corresponding point to the longitudinal axis of the tunnel, and the rotation angle represents the angular position of the corresponding point on the cross section.

[0033] Radial distance The calculation formula is:

[0034] in ， is the radial distance, x is the horizontal coordinate of the original 3D point cloud data, y is the vertical coordinate of the original 3D point cloud data.

[0035] Rotation angle The calculation formula is:

[0036] in ， is the rotation angle, x is the horizontal coordinate of the original 3D point cloud data, y is the vertical coordinate of the original 3D point cloud data.

[0037] Step S22 , expanding the points in the two-dimensional polar coordinates into the two-dimensional plane, so as to map the three-dimensional points of the three-dimensional original point cloud data into the two-dimensional plane coordinates to obtain the two-dimensional plane point cloud data.

[0038] Specifically, such as Figure 3 As shown, the rotation angle in two-dimensional polar coordinates is Corresponding to the horizontal axis of the plane, the radial distance in the two-dimensional polar coordinates The horizontal axis and vertical axis of the corresponding plane. For the coordinates of point P in the three-dimensional original point cloud data, x,y,z ), the corresponding two-dimensional polar coordinate expansion point The coordinates of , ).

[0039] Project each point in the 3D original point cloud data into the 2D polar coordinate system to obtain the corresponding 2D coordinate point ( , ), the points in the polar coordinate system are expanded to a two-dimensional plane to form two-dimensional plane point cloud data, and then the three-dimensional original point cloud data distributed in the three-dimensional space is converted into two-dimensional plane point cloud data in the two-dimensional space. Each point in the two-dimensional plane point cloud data corresponds to a point in the original three-dimensional point cloud, and the spatial topological relationship of the point cloud is retained.

[0040] Due to the rotation angle For example, 0° is equivalent to 360°. The point clouds of adjacent cross sections in the expanded two-dimensional plane point cloud are naturally connected on the horizontal axis, avoiding the jump break caused by occlusion. Figure 3 As shown, the two-dimensional plane point cloud data is unfolded, and its horizontal axis is the rotation angle , the vertical axis is the radial distance , completely retaining the axial consistency of the original three-dimensional point cloud data.

[0041] In addition, by converting the original three-dimensional point cloud data into two-dimensional plane point cloud data, the complexity of the data is reduced, making subsequent processing more efficient. In addition, during the unfolding projection process, the spatial topological relationship of the point cloud is preserved, and the relative position relationship between adjacent points can still be reflected on the two-dimensional plane. At the same time, the processing of two-dimensional plane point clouds is usually easier to implement than three-dimensional point clouds.

[0042] Step S3: Perform Delaunay triangulation on the two-dimensional plane point cloud data to construct a two-dimensional triangulated network.

[0043] Specifically, the steps of performing Delaunay triangulation on the two-dimensional plane point cloud data to obtain a two-dimensional triangulated network include: Step S31 : pre-processing the two-dimensional plane point cloud data, including removing noise points and abnormal points, so as to improve the quality and efficiency of triangulation.

[0044] Step S32: Select three non-collinear point coordinates in the two-dimensional plane point cloud data to construct an initial triangle.

[0045] Specifically, the coordinates of the three non-collinear points are selected to be located at the edge or key position of the two-dimensional plane point cloud data as much as possible to ensure the stability and coverage of the subsequent triangulation process.

[0046] Step S33, using the divide-and-conquer algorithm to process large-scale two-dimensional point clouds, the time complexity is , which is superior to the point-by-point interpolation method. Local density compensation is performed on the point cloud in the fault zone area. Linear interpolation is used to generate virtual points based on the density of neighboring points. The remaining point cloud points are then gradually interpolated according to the empty circle property of Delaunay triangulation. For each point to be interpolated, the surrounding triangles are searched and checked to see if the point satisfies the empty circle property. If not, the topology of the triangulated network is adjusted through operations such as edge flipping until the empty circle property is satisfied, generating new triangles. This ensures that the 2D triangulated network can still generate a continuous mesh in low-density areas.

[0047] Specifically, such as Figure 4As shown in the figure, the core of Delaunay triangulation is to ensure that the circumcircle of any triangle contains no other data points, thereby maximizing the minimum angle and avoiding the generation of narrow triangles. In a two-dimensional plane, point clouds are triangulated using point-by-point interpolation or a divide-and-conquer algorithm. Because the two-dimensional point cloud retains the axial consistency and topological relationships of the original three-dimensional point cloud, the triangulation network can still generate a continuous and uniform mesh in the fault zone through global optimization.

[0048] Step S4: Mark each vertex of the two-dimensional triangulation network at a unique corresponding point in the three-dimensional point cloud data according to vertex association.

[0049] Specifically, in step S2, when the three-dimensional original point cloud data is converted into two-dimensional plane point cloud data through unfolding projection, each point in the two-dimensional plane point cloud data has a corresponding two-dimensional coordinate ( ) and the three-dimensional coordinate points of the three-dimensional original point cloud data ( x,y,z Therefore, after the 2D triangulation is completed, a one-to-one correspondence is established between the 2D triangulation vertices and the 3D coordinate points of the 3D original point cloud data, and each vertex of the 2D triangulation is marked with a unique 3D coordinate point in the 3D original point cloud data ( x,y,z ).

[0050] Step S5: Map the topological relationship of the two-dimensional triangulated network to the three-dimensional original point cloud, and directly convert the two-dimensional adjacent edges into three-dimensional adjacent edges through the edge mapping rule to generate a three-dimensional surface model of the tunnel.

[0051] Specifically, the steps of mapping the topological relationship of the two-dimensional triangulated network to the three-dimensional original point cloud, directly converting the two-dimensional adjacent edges into three-dimensional adjacent edges through the edge mapping rule, and generating the three-dimensional surface model of the tunnel include: Step S51: converting adjacent edges of the two-dimensional triangulated mesh into three-dimensional adjacent edges.

[0052] Specifically, for each edge connecting two vertices in a 2D triangulated network, the mapping table is used to find the two 3D coordinate points in the original 3D point cloud data corresponding to these two vertices. These two 3D coordinate points are then directly connected in 3D space to form a 3D edge. This preserves the topological structure of the 2D plane and ensures that the spatial relationship between adjacent points in the 3D triangulated network is consistent with the 2D triangulated network plane.

[0053] Step S52: combine all three-dimensional adjacent edges to generate a closed triangular mesh surface, and then generate a three-dimensional surface model of the tunnel.

[0054] Specifically, the algorithm traverses all triangular patches in the 2D triangulated mesh, maps the three vertices of each patch to 3D space, and connects them in the same order to form 3D triangles. The resulting 3D triangulated mesh covers the entire tunnel surface, effectively filling in areas with occlusion caused by faults and avoiding isolated triangulations and open holes.

[0055] The present invention provides a method for reconstructing the surface of a mobile measurement point cloud based on a tunnel scene, comprising: using a mobile measurement device to obtain three-dimensional original point cloud data in the tunnel scene; converting the three-dimensional original point cloud data into two-dimensional plane point cloud data through unfolding and projection; performing Delaunay triangulation on the two-dimensional plane point cloud data to construct a two-dimensional triangulated network; marking each vertex of the two-dimensional triangulated network with a unique corresponding point in the three-dimensional point cloud data according to vertex association; mapping the topological relationship of the two-dimensional triangulated network to the three-dimensional original point cloud, and directly converting the two-dimensional adjacent edges into three-dimensional adjacent edges through edge mapping rules to generate a three-dimensional surface model of the tunnel. The present invention provides a method for reconstructing the surface of a mobile measurement point cloud based on a tunnel scene, comprising: obtaining three-dimensional original point cloud data in the tunnel scene through unfolding and projection, converting the three-dimensional original point cloud into a two-dimensional plane point cloud, and then performing Delaunay triangulation, thereby reducing the computational complexity of the network construction from about O( ) is reduced to a two-dimensional network , greatly reducing the time required for tunnel point cloud construction; utilizing the fully connected characteristics of the two-dimensional plane, the proportion of isolated triangulated networks and non-closed holes is close to zero, effectively filling the fracture zone area caused by occlusion, and avoiding isolated triangulated networks and non-closed holes; in view of the geometric closure and scanning center fixity of the tunnel scene, a point cloud geometry reconstruction method based on cross-dimensional topological relationship transmission is proposed, which conforms to the characteristics of tunnel point cloud data and has high practicality and adaptability.

[0056] Second embodiment: like Figure 5 As shown, the second embodiment of the present invention provides a mobile measurement point cloud surface reconstruction system based on a tunnel scene, the system comprising: a three-dimensional data acquisition module 201, a data projection conversion module 202, a two-dimensional triangulation construction module 203, a vertex association module 204, and a mapping output module 205.

[0057] Specifically, the three-dimensional data acquisition module 201 is used to use a mobile measurement device to obtain three-dimensional original point cloud data in the tunnel scene; the data projection conversion module 202 is used to convert the three-dimensional original point cloud data into two-dimensional plane point cloud data through unfolding projection; the two-dimensional triangulation network construction module 203 is used to perform Delaunay triangulation on the two-dimensional plane point cloud data to construct a two-dimensional triangulation network; the vertex association module 204 is used to mark the vertex of each two-dimensional triangulation network with a unique corresponding point in the three-dimensional point cloud data based on vertex association; the mapping output module 205 is used to map the topological relationship of the two-dimensional triangulation network to the three-dimensional original point cloud, and directly convert the two-dimensional adjacent edges into three-dimensional adjacent edges through the edge mapping rule to generate a three-dimensional surface model of the tunnel.

[0058] It is not difficult to find that this embodiment is a system embodiment corresponding to the first embodiment, and this embodiment can be implemented in conjunction with the first embodiment. The relevant technical details mentioned in the first embodiment are still valid in this embodiment, and to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied to the first embodiment.

[0059] It is worth noting that all modules involved in this embodiment are logical modules. In actual applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, to highlight the innovations of the present invention, this embodiment does not include units that are not closely related to solving the technical problems proposed by the present invention. However, this does not mean that other units do not exist in this embodiment.

[0060] The third embodiment of the present invention relates to a network side server, such as Figure 6 As shown, it includes at least one processor 302; and a memory 301 that is communicatively connected to the at least one processor 302; wherein the memory 301 stores instructions that can be executed by the at least one processor 302, and the instructions are executed by the at least one processor 302 to enable the at least one processor 302 to execute the above-mentioned data processing method.

[0061] Memory 301 and processor 302 are connected using a bus. The bus can include any number of interconnected buses and bridges, connecting various circuits of one or more processors 302 and memory 301. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits. These are all well known in the art and are therefore not described further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single component or multiple components, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by processor 302 is transmitted over a wireless medium via an antenna. Furthermore, the antenna receives data and transmits it to processor 302.

[0062] The processor 302 is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. The memory 301 can be used to store data used by the processor 302 when performing operations.

[0063] A fourth embodiment of the present invention relates to a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method for surface reconstruction of a mobile measurement point cloud based on a tunnel scenario in the first embodiment.

[0064] That is, those skilled in the art will understand that all or part of the steps in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a program. The program is stored in a storage medium and includes a number of instructions for causing a device (which may be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes: a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc., various media that can store program code.

[0065] The above is only an embodiment of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme is not described in detail here. Ordinary technicians in the field are aware of all common technical knowledge in the technical field of the invention before the application date or priority date, can obtain all existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the field can improve and implement this scheme in combination with their own abilities under the inspiration given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several variations and improvements can be made, which should also be regarded as the scope of protection of the present invention. These will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

[0066] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for surface reconstruction of mobile measurement point cloud based on tunnel scene, characterized in that: include: S1, using mobile measurement equipment to obtain 3D raw point cloud data in the tunnel scene; S2, converting the 3D original point cloud data into 2D plane point cloud data through unfolding projection; S3, performing Delaunay triangulation on the two-dimensional plane point cloud data to construct a two-dimensional triangulated network; S4, based on vertex association, marking each vertex of the 2D triangulated network at a unique corresponding point in the 3D point cloud data; S5, maps the topological relationship of the 2D triangulated network to the 3D original point cloud, and directly converts the 2D adjacent edges into 3D adjacent edges through the edge mapping rule to generate the 3D surface model of the tunnel.

2. The method for surface reconstruction of mobile measurement point cloud based on tunnel scene according to claim 1 is characterized in that: The three-dimensional original point cloud data of the tunnel scene is obtained using a mobile measuring device equipped with a laser radar. The laser radar of the mobile measuring device scans along the longitudinal axis of the tunnel, and the cross section is orthogonal to the longitudinal axis of the tunnel, so that the three-dimensional original point cloud distribution has axial continuity. The laser radar emits laser pulses and measures the time difference of the reflected light to determine the distance, thereby obtaining a large amount of three-dimensional original point cloud data.

3. The method for surface reconstruction of mobile measurement point cloud based on tunnel scene according to claim 1, characterized in that: The converting of the three-dimensional original point cloud data into two-dimensional plane point cloud data by unfolding and projecting includes: Step S21, converting the three-dimensional original point cloud data into two-dimensional polar coordinates; Step S22 , expanding the points in the two-dimensional polar coordinates into the two-dimensional plane, so as to map the three-dimensional points of the three-dimensional original point cloud data into the two-dimensional plane coordinates to obtain the two-dimensional plane point cloud data.

4. The method for surface reconstruction of mobile measurement point cloud based on tunnel scene according to claim 3 is characterized in that: The process of converting the three-dimensional original point cloud data into two-dimensional polar coordinates includes: The three-dimensional original point cloud data uses the tunnel longitudinal axis as the reference line to establish the global coordinate system as follows: x,y,z ),in, z The axis corresponds to the longitudinal axis of the tunnel, and the cross section is located at xy On a plane; Project each point in the 3D original point cloud data into the polar coordinate system ( ), is the rotation angle, is the radial distance; Radial distance The calculation formula is: in ， is the radial distance, x is the horizontal coordinate of the original 3D point cloud data, y is the vertical coordinate of the original 3D point cloud data; Rotation angle The calculation formula is: in ， is the rotation angle, x is the horizontal coordinate of the original 3D point cloud data, y is the vertical coordinate of the original 3D point cloud data; Expanding the points in the two-dimensional polar coordinates into the two-dimensional plane to map the three-dimensional points of the three-dimensional original point cloud data into the two-dimensional plane coordinates to obtain the two-dimensional plane point cloud data includes: The rotation angle in two-dimensional polar coordinates Corresponding to the horizontal axis of the plane, the radial distance in the two-dimensional polar coordinates The horizontal axis and vertical axis of the corresponding plane, for the coordinates of point P in the three-dimensional original point cloud data, are ( x,y,z ), the corresponding two-dimensional polar coordinate expansion point The coordinates of , ),; Project each point in the 3D original point cloud data into the 2D polar coordinate system to obtain the corresponding 2D coordinate point ( , ), the points in the polar coordinate system are expanded to a two-dimensional plane to form two-dimensional plane point cloud data, so that the three-dimensional original point cloud data distributed in the three-dimensional space is converted into two-dimensional plane point cloud data in the two-dimensional space.

5. The method for surface reconstruction of mobile measurement point cloud based on tunnel scene according to claim 1, characterized in that: The Delaunay triangulation of the two-dimensional plane point cloud data to construct a two-dimensional triangulated network includes: Step S31, preprocessing the two-dimensional plane point cloud data, including removing noise points and abnormal points, to improve the quality and efficiency of triangulation; Step S32, selecting three non-collinear point coordinates in the two-dimensional plane point cloud data to construct an initial triangle; Step S33, using the divide-and-conquer algorithm to process large-scale two-dimensional point clouds, the time complexity is , which is superior to the point-by-point interpolation method, performs local density compensation on the point cloud in the fault zone area, generates virtual points using linear interpolation according to the density of neighboring points, and gradually interpolates the remaining point cloud points according to the empty circle property criterion of Delaunay triangulation.

6. The method for surface reconstruction of mobile measurement point cloud based on tunnel scene according to claim 1, characterized in that: The method of marking each vertex of the two-dimensional triangulated network at a unique corresponding point in the three-dimensional point cloud data according to vertex association includes: when the three-dimensional original point cloud data is converted into two-dimensional plane point cloud data through unfolding projection, each point in the two-dimensional plane point cloud data has a corresponding two-dimensional coordinate ( ) and the three-dimensional coordinate points of the three-dimensional original point cloud data ( x,y,z After the 2D triangulation is completed, a one-to-one correspondence is established between the 2D triangulation vertices and the 3D coordinate points of the 3D original point cloud data, and each vertex of the 2D triangulation is marked with a unique 3D coordinate point in the 3D original point cloud data ( x,y,z ).

7. The method for surface reconstruction of mobile measurement point cloud based on tunnel scene according to claim 1, characterized in that: The topological relationship of the two-dimensional triangulated network is mapped to the three-dimensional original point cloud, and the two-dimensional adjacent edges are directly converted into three-dimensional adjacent edges through the edge mapping rule to generate a three-dimensional surface model of the tunnel, including: Step S51: converting adjacent edges of the two-dimensional triangulated mesh into three-dimensional adjacent edges. Step S52: combine all three-dimensional adjacent edges to generate a closed triangular mesh surface, and then generate a three-dimensional surface model of the tunnel.

8. A mobile measurement point cloud surface reconstruction system based on tunnel scenes, characterized in that: The method for surface reconstruction of a mobile measurement point cloud based on a tunnel scene according to claims 1 to 7 is applied to the system comprising: A three-dimensional data acquisition module is used to acquire three-dimensional original point cloud data in the tunnel scene using a mobile measurement device; The data projection conversion module is used to convert the three-dimensional original point cloud data into two-dimensional plane point cloud data through unfolding projection; 2D triangulation module, used to perform Delaunay triangulation on 2D plane point cloud data to construct 2D triangulation network; A vertex association module is used to mark each vertex of the two-dimensional triangulation network with a unique corresponding point in the three-dimensional point cloud data based on vertex association; The mapping output module is used to map the topological relationship of the two-dimensional triangulated network into the three-dimensional original point cloud, and directly convert the two-dimensional adjacent edges into three-dimensional adjacent edges through the edge mapping rule to generate the three-dimensional surface model of the tunnel.

9. A network-side server, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the mobile measurement point cloud surface reconstruction method based on the tunnel scene according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for surface reconstruction of a mobile measurement point cloud based on a tunnel scene according to any one of claims 1 to 7 is implemented.

Images