Point cloud data registration method, device, equipment and storage medium

Abstract

The application discloses a point cloud data registration method and device, equipment and a storage medium, and belongs to the field of tunnel data processing. A first point cloud data set of a tunnel inner wall surface and geometric features of the tunnel inner wall surface are acquired; the first point cloud data is obtained by scanning the tunnel inner wall surface by a first laser scanning device; a registration point cloud data set is determined from the first point cloud data set according to the geometric features; initial transformation parameters corresponding to the registration point cloud data set are obtained; and a first initial coordinate of the registration point cloud data set is determined according to the initial transformation parameters, so as to obtain a registered first point cloud data set. Thus, the application utilizes the characteristics of the geometric features, such as the plane and regular curved surface, of the tunnel inner wall surface, can improve the registration stability for the scanning point cloud with a large amount of background and noise, and can improve the registration accuracy and speed.

Description

Technical Field

[0001] This application relates to the field of tunnel data processing, and in particular to a point cloud data registration method, apparatus, device, and storage medium. Background Technology

[0002] During tunnel construction, measuring instruments are used to scan and measure the inner walls of the tunnel to ensure the accuracy of the construction.

[0003] Currently, 3D scanners are commonly used to measure tunnel walls due to their advantages such as high data sampling rate, high resolution, high precision, and digital acquisition. However, due to the limited viewing angle of laser scanners and the mutual occlusion between objects, it is very likely that a single-station scan will not obtain all the point cloud data of a large and complex object. Therefore, it is necessary to set up stations in multiple directions and angles to perform multi-station scans on an object in order to obtain complete multi-view point cloud data. However, the point cloud data obtained from different stations are all local coordinate systems of the local station, and their origins and coordinate axes are different. Therefore, the measured point cloud data cannot be regarded as a continuous dataset.

[0004] Application content

[0005] The main objective of this application is to provide a point cloud data registration method, apparatus, device, and storage medium, which aims to solve the technical problem that point cloud data measured at multiple sites cannot be used as a continuous dataset.

[0006] To achieve the above objectives, this application provides a point cloud data registration method, the method comprising:

[0007] A first point cloud dataset and geometric features of the tunnel inner wall surface are obtained; wherein the first point cloud data is obtained by scanning the tunnel inner wall surface with a first laser scanning device;

[0008] Based on the geometric features, a registration point cloud dataset is determined from the first point cloud dataset;

[0009] Obtain the initial transformation parameters corresponding to the registered point cloud dataset;

[0010] Based on the initial transformation parameters, the first initial coordinates of the registered point cloud dataset are determined to obtain the first registered point cloud dataset.

[0011] Optionally, after determining the initial position of the registered point cloud dataset based on the initial transformation parameters to obtain the registered first point cloud dataset, the method further includes:

[0012] A second point cloud dataset is obtained from the inner wall surface of the tunnel; wherein the second point cloud dataset is obtained by scanning the inner wall surface of the tunnel with a second laser scanning device, and the first laser scanning device and the second laser scanning device are spaced apart from each other in the axial direction of the tunnel;

[0013] The preceding transformation parameters are used as the current transformation parameters; the preceding transformation parameters are the initial coordinate transformation parameters between the registered first point cloud dataset and the second point cloud dataset.

[0014] From the registered first point cloud dataset, a first preset point is determined, and a transformation is performed according to the current transformation parameters to obtain the first transformation preset point;

[0015] Select the second preset point from the second point cloud dataset that is closest in physical distance to the first conversion preset point;

[0016] The target conversion parameters are calculated based on the first preset point and the second preset point;

[0017] The first preset point is transformed according to the target transformation parameters to obtain the target point; the target point is in the second point cloud dataset, and the coordinate data of the target point is consistent with the coordinate data of the second preset point.

[0018] Optionally, after calculating the target transformation parameters based on the first preset point and the second preset point, the method further includes:

[0019] Determine whether the difference between the current conversion parameter and the target conversion parameter is less than a preset threshold;

[0020] If so, then the coordinate transformation of the first preset point according to the target transformation parameters is performed to obtain the target point.

[0021] Optionally, after determining whether the difference between the current conversion parameter and the target conversion parameter is less than a preset threshold, the method further includes:

[0022] If not, the target conversion parameter is used as the preceding conversion parameter, the current conversion parameter is updated according to the target conversion parameter, and the process returns to the step of converting the first preset point according to the current conversion parameter to obtain the first conversion preset point, until the difference between the preceding conversion parameter and the target conversion parameter is less than the preset threshold.

[0023] Optionally, the step of transforming the first preset point in the first point cloud dataset according to the current transformation parameters to obtain the first transformation preset point includes:

[0024] The first preset point and the current conversion parameter are converted according to Formula 1 to obtain the first conversion preset point, wherein Formula 1 includes:

[0025] q i =R k-1 p i +T k-1 ,

[0026] Where, q i R is the first conversion preset point. k-1 and T k-1 p is the current conversion parameter. i Let k be the first preset point, k be the iteration number, and k∈[1,n].

[0027] Optionally, the step of calculating the target transformation parameters based on the first preset point and the second preset point includes:

[0028] Based on the first preset point, the second preset point, and Formula 2, the distance between the nearest point pairs is obtained, wherein Formula 2 includes:

[0029] D = R k p i +T k -q i ,

[0030] Among them, R k and T k p is the target transformation parameter. i Let q be the first preset point. i Let k be the second preset point, k be the iteration number, and k∈[1,n];

[0031] The target transformation parameters are obtained by taking the minimum metric of the distance between the nearest point pairs.

[0032] Optionally, calculating the target transformation parameters based on the first preset point and the second preset point includes:

[0033] The target transformation parameters are obtained by taking the minimum metric of the distance between the nearest point pairs according to Formula 3, wherein Formula 3 includes:

[0034]

[0035] Where e is the natural constant, N is a natural number, and R is a natural number. k and T k p is the target transformation parameter. i Let q be the first preset point. i Let k be the second preset point, k be the number of iterations, and k∈[1,n], and min be the minimum value.

[0036] Secondly, this application also provides a point cloud data processing apparatus, the apparatus comprising:

[0037] The acquisition module is used to acquire a first point cloud dataset and geometric features of the tunnel inner wall surface; wherein the first point cloud data is obtained by scanning the tunnel inner wall surface with a first laser scanning device;

[0038] The determination module is used to determine the registration point cloud dataset from the first point cloud dataset based on the geometric features;

[0039] The first acquisition module is used to obtain the initial transformation parameters corresponding to the registration point cloud dataset;

[0040] The second acquisition module is used to determine the first initial coordinates of the registered point cloud dataset based on the initial transformation parameters, so as to obtain the registered first point cloud dataset.

[0041] Thirdly, this application provides a point cloud data processing device, including: a processor, a memory, and a point cloud data registration program stored in the memory, wherein when the point cloud data registration program is executed by the processor, it implements the steps of the point cloud data registration method described above.

[0042] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the point cloud data registration method of any embodiment of this application.

[0043] This application proposes a point cloud data registration method, which involves acquiring a first point cloud dataset of the tunnel inner wall surface and the geometric features of the tunnel inner wall surface. The first point cloud data is obtained by scanning the tunnel inner wall surface using a first laser scanning device. Based on the geometric features, a registration point cloud dataset is determined from the first point cloud dataset. Initial transformation parameters corresponding to the registration point cloud dataset are obtained. Based on the initial transformation parameters, first initial coordinates of the registration point cloud dataset are determined to obtain the registered first point cloud dataset. Therefore, this application utilizes the planar and regular curved surface geometric features of the tunnel inner wall surface to improve registration stability, accuracy, and speed, even for scanned point clouds with significant background and noise. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of the point cloud data processing device of this application;

[0045] Figure 2 This is a flowchart illustrating the first embodiment of the point cloud data registration method of this application;

[0046] Figure 3 This is a flowchart illustrating the second embodiment of the point cloud data registration method of this application;

[0047] Figure 4 This is a flowchart illustrating the third embodiment of the point cloud data registration method of this application;

[0048] Figure 5 This is a schematic diagram of the functional modules of the point cloud data processing device of this application.

[0049] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0050] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0051] Due to the limitations of existing technology, which relies on single-site measurements of the tunnel wall, the resulting point cloud dataset cannot fully represent the measurement data of the object being measured. To obtain complete multi-view point cloud data, multiple stations need to be set up for scanning and measuring the object. However, since the point cloud data acquired by different stations are all local coordinate systems of their respective stations, with different origins and coordinate axes, the point cloud data obtained from different stations are not in the same coordinate system. Therefore, the obtained point cloud data cannot be considered a continuous dataset and cannot be directly used as the point cloud dataset for the tunnel wall.

[0052] This application provides a solution that acquires a first point cloud dataset and geometric features of the tunnel inner wall surface. The first point cloud dataset is obtained by scanning the tunnel inner wall surface using a first laser scanning device. Based on the geometric features, a registration point cloud dataset is determined from the first point cloud dataset. Initial transformation parameters corresponding to the registration point cloud dataset are obtained. Based on the initial transformation parameters, first initial coordinates of the registration point cloud dataset are determined to obtain the registered first point cloud dataset. Therefore, this application utilizes the planar and regular curved surface geometric features of the tunnel inner wall surface to improve registration stability, accuracy, and speed, even with scanned point clouds containing significant background and noise.

[0053] Reference Figure 1 , Figure 1 This is a schematic diagram of the point cloud data processing device in the hardware operating environment involved in the embodiments of this application.

[0054] like Figure 1As shown, the point cloud data processing device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen and an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as a disk drive. The memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001.

[0055] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on the point cloud data processing device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0056] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a data storage module, a network communication module, a user interface module, and a point cloud data registration program.

[0057] exist Figure 1 In the point cloud data processing device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the point cloud data processing device of this application can be set in the point cloud data processing device. The point cloud data processing device calls the point cloud data registration program stored in the memory 1005 through the processor 1001 and executes the point cloud data registration method provided in the embodiment of this application.

[0058] Based on, but not limited to, the hardware structure of the point cloud data processing device described above, this application provides a first embodiment of a point cloud data registration method. (Refer to...) Figure 2 , Figure 2 A flowchart illustrating the first embodiment of the point cloud data registration method is shown.

[0059] It should be noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown here.

[0060] In this embodiment, the point cloud data registration method includes:

[0061] Step S10: Obtain the first point cloud dataset and the geometric features of the tunnel inner wall surface; wherein, the first point cloud dataset is obtained by scanning the tunnel inner wall surface with a first laser scanning device;

[0062] The point cloud data registration method is implemented by a terminal device with display and interactive functions, such as a laptop computer; this application does not limit this. For example, construction workers might import the first point cloud dataset obtained after scanning the inner wall surface of a tunnel using a 3D laser scanner into a laptop computer.

[0063] In this embodiment, point cloud data is a set of points obtained after acquiring the spatial coordinates of each sampling point on the surface of an object; it is also called a massive point set of surface characteristics of the target object. Scanning data is recorded in the form of points, each containing three-dimensional coordinates. Some points may contain color information (RGB) or reflectance intensity information. Therefore, the first point cloud dataset can be a set of multiple vectors in a three-dimensional coordinate system obtained by scanning the tunnel inner wall surface using a first laser scanning device. Geometric features can be the basic elements constituting a building, such as points, lines, planes, and regular curved surface geometric elements that constitute the contour of the tunnel inner wall.

[0064] It is understandable that the features of a building, such as points, lines, and surfaces, all have strict set constraints. Based on the coincidence of planes in the building point cloud, two types of constraints can be established: (1) the condition that a point lies on a plane; and (2) the condition that two normals are parallel to each other. The point cloud data registration method based on the geometric features of building points, lines, and surfaces mainly includes the following constraints: coplanar condition, fixed distance condition, copular vertical line condition, point on a straight line condition, fixed distance condition from a point to a straight line condition, coincidence of two spatial lines condition, coplanar condition of two spatial lines condition, fixed direction condition of a straight line condition, point on a plane condition, and fixed distance condition from a point to a plane condition, etc.

[0065] Step S20: Determine the registration point cloud dataset from the first point cloud dataset based on the geometric features;

[0066] Step S30: Obtain the initial transformation parameters corresponding to the registration point cloud dataset;

[0067] Step S40: Determine the first initial coordinates of the registered point cloud dataset according to the initial transformation parameters, so as to obtain the first registered point cloud dataset.

[0068] Specifically, the registration point cloud dataset can be a set of multiple vectors to be registered, determined from the first point cloud dataset. The initial transformation parameters can be intermediate variables obtained after the initial calculation of the registration point cloud dataset based on coordinate transformation principles. The first initial coordinates are the geographic coordinates of the registered point cloud dataset.

[0069] Specifically, after acquiring the first point cloud dataset, the normal vector and curvature features of the sampling points are calculated based on the neighborhood points of the first point cloud dataset. The obtained curvature features are used as connection features to filter the point cloud datasets that need to be registered, and a set of matching point pairs corresponding to the point cloud datasets to be registered is determined. Then, each matching point is checked and judged according to geometric feature constraints, erroneous point pairs are identified, and valid matching point pairs are obtained, i.e., the registered point cloud dataset. The initial transformation parameters of the registered point cloud dataset are calculated based on the coordinate transformation principle, and the first initial coordinates of the registered point cloud dataset are determined according to the initial transformation parameters to obtain the registered first point cloud dataset.

[0070] In this embodiment, a first point cloud dataset and the geometric features of the tunnel inner wall surface are acquired. The first point cloud dataset is obtained by scanning the tunnel inner wall surface using a first laser scanning device. Based on the geometric features, a registration point cloud dataset is determined from the first point cloud dataset. Initial transformation parameters corresponding to the registration point cloud dataset are obtained. Based on the initial transformation parameters, the first initial coordinates of the registration point cloud dataset are determined to obtain the registered first point cloud dataset. Therefore, this application utilizes the geometric features of the tunnel inner wall surface, such as its planar and regular curved surfaces, to improve registration stability, accuracy, and speed, especially for scanned point clouds with significant background and noise.

[0071] Furthermore, as an example, refer to Figure 3 Based on the above Figure 2 The embodiment shown in this application provides a second embodiment of a point cloud data registration method.

[0072] In this embodiment, after step S40, the method further includes:

[0073] Step S401: Obtain a second point cloud dataset of the inner wall surface of the tunnel; wherein the second point cloud dataset is obtained by scanning the inner wall surface of the tunnel with a second laser scanning device, and the first laser scanning device and the second laser scanning device are spaced apart from each other in the axial direction of the tunnel;

[0074] In this embodiment, the second point cloud dataset can be a collection of multiple vectors in a three-dimensional coordinate system obtained by the second laser device scanning the tunnel inner wall surface. The registered first point cloud dataset is contained within the second point cloud dataset; that is, the scanning area of ​​the first laser device is within the scanning area of ​​the second laser device. The first and second laser scanning devices are scanning devices with different measurement stations and spaced apart from each other, and can be three-dimensional laser scanners.

[0075] Understandably, due to the limitations of laser scanner angles, scanning a large area of ​​tunnel interior walls may not yield complete point cloud data from a single station. Therefore, two laser scanning devices need to be set up in a predetermined area of ​​the tunnel interior wall for multi-directional scanning and measurement. Since the laser scanning devices at different stations have different measurement angles and directions, and use different local coordinate systems, the measured point cloud data cannot be considered a continuous dataset. Therefore, it is necessary to unify these independent coordinate systems into a single coordinate system through coordinate transformation, thereby stitching together the point cloud data collected from different stations to form a complete 3D point cloud model, i.e., registering the collected point cloud data.

[0076] Step S402: Use the previous transformation parameters as the current transformation parameters; the previous transformation parameters are the initial coordinate transformation parameters between the registered first point cloud dataset and the second point cloud dataset;

[0077] In this embodiment, the pre-transformation parameter is an intermediate variable that can be filled with other data. In this embodiment, the pre-transformation parameter is the initial coordinate transformation parameter. The current transformation parameter is the coordinate parameter used during this coordinate transformation. The initial coordinate transformation parameter is pre-set parameter data, denoted as R. 0 and T 0 .

[0078] Step S403: Determine a first preset point from the registered first point cloud dataset, and perform conversion according to the current conversion parameters to obtain the first conversion preset point;

[0079] In this embodiment of the application, the first preset point can be any point in the registered first point cloud dataset, denoted as p. i Where i∈[1,n]. The first transformation preset point is the point in the second point cloud dataset that is the same as p. i The corresponding point is denoted as q. i .

[0080] Specifically, the first preset point and the current conversion parameters are converted according to Formula 1 to obtain the first conversion preset point, wherein Formula 1 includes:

[0081] q i =Rk-1 p i +T k-1 ,

[0082] Where, q i R is the first preset conversion point. k-1 and T k-1 p is the current conversion parameter. i Let k be the first preset point, k be the iteration number, and k∈[1,n].

[0083] Set the first preset point p i Based on the initial coordinate transformation parameters, perform coordinate transformation to obtain the coordinates relative to the first preset point p. i The corresponding first conversion preset point q i For example, when i is 1, that is, during the first coordinate transformation, the first preset transformation point is: q1 = R 0 p1+T 0 .

[0084] Step S404: Select the second preset point in the second point cloud dataset that is closest in physical distance to the first conversion preset point;

[0085] It is understandable that the first preset point is a point in the second point cloud dataset. The point that is physically closest to the first preset point is selected as the second preset point. The coordinate data of the second preset point is close to that of the first preset point, and they can be approximated as the same point. Therefore, the second preset point can also be denoted as q. i .

[0086] Step S405: Calculate the target conversion parameters based on the first preset point and the second preset point;

[0087] Step S406: Transform the first preset point according to the target transformation parameters to obtain the target point; the target point is in the second point cloud dataset, and the coordinate data of the target point is consistent with the coordinate data of the second preset point.

[0088] In this embodiment, the target transformation parameter is the parameter for coordinate transformation between the first preset point and the second preset point. The target point is the point in the second point cloud dataset corresponding to the first preset point. The calculated target transformation parameter is more accurate than the initial coordinate transformation parameter.

[0089] It should be understood that in this embodiment, the points in the first point cloud dataset after registration are converted to the second point cloud dataset for representation, so as to stitch the first point cloud dataset and the second point cloud dataset after registration to obtain the complete point cloud data of the tunnel inner wall.

[0090] This embodiment acquires a registered first point cloud dataset and a second point cloud dataset from the tunnel inner wall. A pre-conversion parameter is used as the current conversion parameter, which is the initial coordinate transformation parameter between the registered first and second point cloud datasets. A first preset point in the registered first point cloud dataset is transformed according to the current conversion parameter to obtain a first preset conversion point. A second preset point with the closest physical distance to the first preset conversion point is selected from the second point cloud dataset. Based on the first and second preset points, a target conversion parameter is calculated. The first preset point is then transformed according to the target conversion parameter to obtain the target point. This achieves the stitching of point cloud data measured at different stations, ensuring the continuity of data measured from the tunnel inner wall.

[0091] Furthermore, as an example, refer to Figure 4 Based on the above Figure 3 The embodiment shown in this application provides a third embodiment of a point cloud data registration method.

[0092] In this embodiment, after step S405, the method further includes:

[0093] Step S4051: Determine whether the difference between the current conversion parameter and the target conversion parameter is less than a preset threshold;

[0094] Step S4052: If yes, then perform the coordinate transformation of the first preset point according to the target transformation parameters to obtain the target point.

[0095] In the embodiments of this application, the preset preset can be a threshold used to limit the difference between the pre-conversion parameter and the target conversion parameter.

[0096] Specifically, suppose there is a target transformation parameter R. k and T k The first preset point is transformed according to the first target transformation parameter to obtain the result that is the same as the first preset point p. i The corresponding second preset point q in the second point cloud dataset i Subtracting the two values ​​yields the distance between the closest points, as shown in Formula 2. Formula 2 includes:

[0097] D = R k p i +T k -q i ,

[0098] Among them, R k and T k p is the target transformation parameter. i Let q be the first preset point. i Let k be the second preset point, k be the iteration number, and k∈[1,n].

[0099] Understandably, by taking the minimum metric for the distance between the nearest points, we can determine the first preset point and the second preset point that are infinitely close after the transformation, and then calculate the target transformation parameters at this point, as shown in Formula 3. Formula 3 includes:

[0100]

[0101] Where e is the natural constant, N is a natural number, and R is a natural number. k and T k p is the target transformation parameter. i Let q be the first preset point. i Let k be the second preset point, k be the number of iterations, and k∈[1,n], and min be the minimum value.

[0102] In one example, the difference between the calculated target transformation parameter and the preceding transformation parameter is determined to be less than a preset threshold. The purpose is to select transformation parameters that make the first preset point and the second preset point nearly the same after transformation. The preceding transformation parameter can be the coordinate parameter used in the previous coordinate transformation. In this step, the difference between the target transformation parameter calculated in this coordinate transformation and the preceding transformation parameter is calculated. If the difference is less than the threshold, it means that the target transformation parameter is qualified and can be used directly as the coordinate transformation parameter. If the difference is greater than the threshold, it means that the target transformation parameter calculated in this step has a problem with low accuracy.

[0103] In another example, if the difference is greater than the preset threshold, it means that the target conversion parameter is not qualified. The target conversion parameter is recorded as A, and the target conversion parameter A is used as the pre-conversion parameter. The current conversion parameter is updated according to the target conversion parameter A, and the target conversion parameter is recalculated to obtain a new target conversion parameter, which is recorded as target conversion parameter B. Then it is judged whether the difference between target conversion parameter A and target conversion parameter B is greater than the preset threshold, until the judgment condition is met.

[0104] This embodiment determines whether the difference between the preceding transformation parameter and the target transformation parameter is less than a preset threshold. If so, it performs coordinate transformation on the first preset point according to the target transformation parameter to obtain the target point. If not, it uses the target transformation parameter as the preceding transformation parameter, updates the current transformation parameter according to the target transformation parameter, and returns to perform the step of transforming the first preset point according to the current transformation parameter to obtain the transformed preset point. This process continues until the difference between the preceding transformation parameter and the target transformation parameter is less than the preset threshold. In other words, this embodiment improves the accuracy and precision of coordinate transformation by iteratively judging the transformation parameters of two adjacent coordinate transformations to select transformation parameters that make the transformed first preset point and second preset point nearly identical.

[0105] Based on the same inventive concept, this application provides a point cloud data processing device, referring to... Figure 5 , Figure 5 This is a schematic diagram of the modules of the first embodiment of the point cloud data processing device of this application.

[0106] The acquisition module 10 is used to acquire a first point cloud dataset and geometric features of the tunnel inner wall surface; wherein, the first point cloud data is obtained by scanning the tunnel inner wall surface by a first laser scanning device;

[0107] The determining module 20 is used to determine the registration point cloud dataset from the first point cloud dataset based on the geometric features;

[0108] The first acquisition module 30 is used to obtain the initial transformation parameters corresponding to the registration point cloud dataset;

[0109] The second obtaining module 40 is used to determine the first initial coordinates of the registered point cloud dataset according to the initial transformation parameters, so as to obtain the registered first point cloud dataset.

[0110] It should be noted that the various implementation methods of the point cloud data processing device in this embodiment and the technical effects they achieve can be referred to the various implementation methods of the point cloud data registration method in the foregoing embodiments, and will not be repeated here.

[0111] In this embodiment, through the cooperation of various functional modules, a first point cloud dataset and the geometric features of the tunnel inner wall surface are acquired. The first point cloud data is obtained by scanning the tunnel inner wall surface using a first laser scanning device. Based on the geometric features, a registration point cloud dataset is determined from the first point cloud dataset. Initial transformation parameters corresponding to the registration point cloud dataset are obtained. Based on the initial transformation parameters, the first initial coordinates of the registration point cloud dataset are determined to obtain the registered first point cloud dataset. Therefore, this application utilizes the geometric features of the tunnel inner wall surface, such as its planar and regular curved surfaces, to improve registration stability, accuracy, and speed, even for scanned point clouds with significant background and noise.

[0112] Furthermore, embodiments of this application also propose a computer storage medium storing a point cloud data registration program. When executed by a processor, the point cloud data registration program implements the steps of the point cloud data registration method described above. Therefore, it will not be repeated here. Additionally, the beneficial effects of using the same method will not be repeated here either. For technical details not disclosed in the computer-readable storage medium embodiments of this application, please refer to the description of the method embodiments of this application. As an example, program instructions can be deployed to execute on a single computing device, or on multiple computing devices located at one location, or on multiple computing devices distributed across multiple locations and interconnected via a communication network.

[0113] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0114] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided in this application, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0115] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, and of course, it can also be implemented by special hardware including application-specific integrated circuits, special CPUs, special memory, special components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0116] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A point cloud data registration method, characterized in that, The method includes: A first point cloud dataset and geometric features of the tunnel inner wall surface are obtained; wherein the first point cloud dataset is obtained by scanning the tunnel inner wall surface with a first laser scanning device; Based on the geometric features, a registration point cloud dataset is determined from the first point cloud dataset; Obtain the initial transformation parameters corresponding to the registered point cloud dataset; Based on the initial transformation parameters, determine the first initial coordinates of the registered point cloud dataset to obtain the first registered point cloud dataset; A second point cloud dataset is obtained from the inner wall surface of the tunnel; wherein the second point cloud dataset is obtained by scanning the inner wall surface of the tunnel with a second laser scanning device, and the first laser scanning device and the second laser scanning device are spaced apart from each other in the axial direction of the tunnel; The preceding transformation parameters are used as the current transformation parameters; the preceding transformation parameters are the initial coordinate transformation parameters between the registered first point cloud dataset and the second point cloud dataset. From the registered first point cloud dataset, a first preset point is determined, and a transformation is performed according to the current transformation parameters to obtain the first transformation preset point; Select the second preset point from the second point cloud dataset that is closest in physical distance to the first conversion preset point; The target conversion parameters are calculated based on the first preset point and the second preset point; The first preset point is transformed according to the target transformation parameters to obtain the target point; the target point is in the second point cloud dataset, and the coordinate data of the target point is consistent with the coordinate data of the second preset point.

2. The point cloud data registration method according to claim 1, characterized in that, After calculating the target transformation parameters based on the first preset point and the second preset point, the method further includes: Determine whether the difference between the current conversion parameter and the target conversion parameter is less than a preset threshold; If so, then the coordinate transformation of the first preset point according to the target transformation parameters is performed to obtain the target point.

3. The point cloud data registration method according to claim 2, characterized in that, After determining whether the difference between the current conversion parameter and the target conversion parameter is less than a preset threshold, the method further includes: If not, the target conversion parameter is used as the preceding conversion parameter, the current conversion parameter is updated according to the target conversion parameter, and the process returns to the step of converting the first preset point according to the current conversion parameter to obtain the first conversion preset point, until the difference between the preceding conversion parameter and the target conversion parameter is less than the preset threshold.

4. The point cloud data registration method according to claim 1, characterized in that, The step of transforming the first preset point in the first point cloud dataset according to the current transformation parameters to obtain the first transformation preset point includes: The first preset point and the current conversion parameter are converted according to Formula 1 to obtain the first conversion preset point, wherein Formula 1 includes: ， in, The first conversion preset point, and The current conversion parameter, For the first preset point, Let be the number of iterations, and .

5. The point cloud data registration method according to claim 1, characterized in that, The step of calculating the target conversion parameters based on the first preset point and the second preset point includes: Based on the first preset point, the second preset point, and Formula 2, the distance between the nearest point pairs is obtained, wherein Formula 2 includes: ， in, and The target transformation parameters, For the first preset point, This is the second preset point. Let be the number of iterations, and ; The target transformation parameters are obtained by taking the minimum metric of the distance between the nearest point pairs.

6. A point cloud data processing device, characterized in that, The device includes: The acquisition module is used to acquire a first point cloud dataset and geometric features of the tunnel inner wall surface; wherein, the first point cloud dataset is obtained by scanning the tunnel inner wall surface with a first laser scanning device; The determination module is used to determine the registration point cloud dataset from the first point cloud dataset based on the geometric features; The first acquisition module is used to obtain the initial transformation parameters corresponding to the registration point cloud dataset; The second obtaining module is used to determine the first initial coordinates of the registered point cloud dataset based on the initial transformation parameters, so as to obtain the registered first point cloud dataset. The device is further configured to acquire a second point cloud dataset of the tunnel inner wall surface; wherein the second point cloud dataset is obtained by scanning the tunnel inner wall surface with a second laser scanning device, and the first laser scanning device and the second laser scanning device are spaced apart from each other along the axial direction of the tunnel; a pre-conversion parameter is used as the current conversion parameter; the pre-conversion parameter is the initial coordinate conversion parameter between the registered first point cloud dataset and the second point cloud dataset; a first preset point is determined from the registered first point cloud dataset, and converted according to the current conversion parameter to obtain a first conversion preset point; a second preset point is selected from the second point cloud dataset that is closest in physical distance to the first conversion preset point; a target conversion parameter is calculated based on the first preset point and the second preset point; the first preset point is converted according to the target conversion parameter to obtain a target point; the target point is in the second point cloud dataset, and the coordinate data of the target point is consistent with the coordinate data of the second preset point.

7. A point cloud data processing device, characterized in that, include: A processor, a memory, and a point cloud data registration program stored in the memory, wherein the point cloud data registration program is executed by the processor to implement the steps of the point cloud data registration method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a point cloud data registration program, which, when executed by a processor, implements the point cloud data registration method as described in any one of claims 1 to 5.

Images