Roadbed intelligent construction three-dimensional visualization operation control method and system

Abstract

The invention relates to the technical field of three-dimensional visualization, and discloses a roadbed intelligent construction three-dimensional visualization operation control method and system, and the method comprises the steps: collecting multi-source roadbed data, and carrying out the position correction of the multi-source roadbed data through an improved SLAM algorithm; performing multi-source fusion on the multi-source roadbed data after position correction to obtain roadbed fusion data; constructing a roadbed construction three-dimensional model based on the roadbed fusion data; and carrying out lightweight processing and dynamic operation rendering on the roadbed construction three-dimensional model by adopting a secondary error measurement grid simplification algorithm. According to the method, the multi-dimensional feature entropy of different roadbed areas is calculated, the roadbed areas with complex terrains are screened for multiple times of resampling, the three-dimensional modeling accuracy of the complex roadbed areas is improved, triangular surface vertex merging is performed based on the error matrix of the triangular surface vertexes in the roadbed construction three-dimensional model, lightweight processing of the roadbed construction three-dimensional model is realized, and the construction efficiency of the roadbed construction three-dimensional model is improved. And carrying out real-time three-dimensional visualization on the roadbed construction operation process.

Description

Technical Field

[0001] The present invention relates to the field of three-dimensional visualization, and in particular to a three-dimensional visualization operation control method and system for intelligent construction of roadbed. Background Art

[0002] With the rapid development of social economy and the increase in transportation demand, highway construction has become an important part of national infrastructure construction. Roadbed is the foundation of highway construction, and its quality and stability directly affect the service life and safety of the highway. In the traditional roadbed construction process, due to the complex construction environment, many operation links, and the difficulty of quality control, it is often difficult to achieve refined management, and various problems such as substandard quality and long construction period are prone to occur during the construction process. Therefore, how to improve the quality, efficiency and accuracy of roadbed construction has become a major challenge in current road construction. With the development of information technology and intelligent technology, three-dimensional visualization technology and operation control methods are increasingly being applied to engineering construction, especially in the intelligent construction of roadbed. Three-dimensional visualization technology integrates, displays and analyzes various links, operation processes and data of roadbed construction through digital means, making the monitoring and management of the construction process more intuitive and transparent. At the same time, the rise of intelligent construction enables the construction process to use automation and information technology to perceive and control the construction environment and progress in real time, thereby improving the accuracy and safety of construction. However, since roadbed engineering involves large-scale terrain and structural models, it faces serious model distortion and loading jams in complex geological scenes. Summary of the invention

[0003] In view of this, the present invention proposes a path intelligent construction three-dimensional visualization operation control method and system, which realizes real-time visualization of the three-dimensional model by optimizing the three-dimensional model rendering process and accuracy, thereby achieving the purpose of precise quality management and real-time monitoring.

[0004] To achieve the above object, the present invention provides a three-dimensional visualization operation control method for intelligent construction of roadbed, comprising the following steps: S1: Collect multi-source roadbed data, use the improved SLAM algorithm to perform position correction on the multi-source roadbed data, and obtain the multi-source roadbed data after position correction; S2: Perform multi-source fusion on the multi-source roadbed data after position correction to obtain roadbed fusion data; S3: Construct a 3D model of roadbed construction based on roadbed fusion data; S4: Use the quadratic error measure grid simplification algorithm to lightweight the roadbed construction three-dimensional model to obtain a lightweight roadbed construction three-dimensional model; S5: Dynamically render the 3D model of lightweight roadbed construction and visualize the roadbed construction process.

[0005] As a further improvement method of the present invention: Optionally, the three-dimensional point cloud data includes three-dimensional position coordinates and a color vector; The radar echo data includes three-dimensional position coordinates, signal amplitude, phase information, interference phase and coherence; The satellite positioning data includes three-dimensional position coordinates and movement speed; The multi-source roadbed data is represented in the form of: ; ; in: G represents multi-source roadbed data, Represents a 3D point cloud data set, represents a radar echo data set, represents the satellite positioning data set, and N represents the number of data collection; Represents a 3D point cloud data set The nth group of 3D point cloud data in Represents a radar echo data set The nth group of radar echo data in Represents a collection of satellite positioning data The nth group of satellite positioning data in; An improved SLAM algorithm is used to construct a multi-source factor error function of the multi-source roadbed data. The multi-source factor error function takes a global position correction parameter as a variable and the degree of matching of the three-dimensional position coordinates of the multi-source roadbed data after the global position correction as a function value. The improved SLAM algorithm is used to perform position matching on the multi-source roadbed data, and takes the three-dimensional position coordinates in the three-dimensional point cloud data, radar echo data and satellite positioning data as nodes and the multi-source factor error functions between the nodes as edges to construct a factor graph.

[0006] Optionally, the expression of the multi-source factor error function is: ; ; ; ; ; ; in: Indicates The quality weight of the error factor, represents the global position correction parameter to be solved, It represents the position correction parameters of 3D point cloud data, radar echo data and satellite positioning data respectively. Indicates The first to third factor errors are the factor errors of three-dimensional point cloud data, radar echo data and satellite positioning data, respectively. The first to third factor error functions are the factor error functions of three-dimensional point cloud data, radar echo data and satellite positioning data, respectively. express The standard deviation of the N three-dimensional position coordinates in ; represents the L1 norm, Represents 3D point cloud data , radar echo data And satellite positioning data The three-dimensional position coordinates in ; Representing a collection Zhongyu The nearest three-dimensional position coordinates; Representing a collection Zhongyu The nearest three-dimensional position coordinates; Representing a collection Zhongyu The nearest 3D position coordinates.

[0007] Optionally, the multi-source factor error function is solved with the goal of minimizing the multi-source factor error function to obtain a global position correction parameter, including: Initializing and generating global position correction parameters, and using the generated global position correction parameters as input values ​​of the multi-source factor error function; Calculate the first-order gradient of the multi-source factor error function as the Jacobian matrix of the generated global position correction parameters, and calculate the second-order inverse of the Jacobian matrix as the Hessian matrix of the generated global position correction parameters; Iterating the generated global position correction parameters based on the Jacobian matrix and the Hessian matrix, and re-iterating the Jacobian matrix and the Hessian matrix of the global position correction parameters; After each iteration, the rate of change of the global position correction parameter is calculated. If the rate of change is less than a preset threshold, the iteration is terminated and the current iteration result is used as the solved global position correction parameter; the three-dimensional position coordinates in the multi-source roadbed data are corrected using the global position correction parameter to obtain the multi-source roadbed data after position correction.

[0008] Optionally, performing multi-source fusion on the multi-source roadbed data after position correction includes: Calculating the three-dimensional position coordinate distances among the three-dimensional point cloud data, radar echo data, and satellite positioning data in the multi-source roadbed data after position correction, and fusing the data from different sources whose three-dimensional position coordinate distances are less than a preset distance threshold as multi-source fused roadbed data under the same three-dimensional position coordinates; The multi-dimensional characteristic entropy of different roadbed areas is calculated based on the multi-source fused roadbed data. The larger the multi-dimensional characteristic entropy, the more complex the surface terrain. Multiple multi-source roadbed data collection is performed in the roadbed area where the multi-dimensional characteristic entropy is higher than the preset entropy threshold. Based on the global position correction parameters, the three-dimensional position coordinates in the newly collected multi-source roadbed data are corrected and integrated into the multi-source fused roadbed data to form roadbed fusion data.

[0009] Optionally, the three-dimensional model of roadbed construction includes a filtering module and a three-dimensional surface reconstruction module, the filtering module is used to extract multi-source information of each group of multi-source fused roadbed data in the roadbed fusion data, and perform multi-scale filtering on the multi-source fused roadbed data to obtain filtered roadbed fusion data, and the three-dimensional surface reconstruction module is used to extract three-dimensional position coordinates and multi-source information from the filtered roadbed fusion data, and perform three-dimensional network reconstruction using a Poisson reconstruction method to obtain a three-dimensional model of roadbed construction; The multi-source information in the multi-source fusion roadbed data is the signal amplitude, phase information, interference phase, coherence and movement speed in the multi-source fusion roadbed data; The multi-scale filtering processing method is as follows: based on multi-source information, filter scale information of multi-source fused roadbed data is calculated; if the filter scale information is higher than a preset filter threshold, it indicates that the multi-source fused roadbed data is noise data; the noise data in the roadbed fusion data is filtered and cleaned to obtain filtered roadbed fusion data; The Poisson reconstruction method is: using principal component analysis to calculate the normal vector of each group of multi-source fused roadbed data in the filtered roadbed fusion data, wherein the higher the coherence, the higher the weight of the multi-source fused roadbed data in the normal vector calculation process; Constructing an octree to store the multi-source fused roadbed data in the filtered roadbed fusion data, and performing spatial hierarchical division on the filtered roadbed fusion data; Poisson equations at different spatial levels are constructed and solved to obtain smooth implicit surface equations of different roadbed areas. Isosurfaces are extracted to generate closed triangular meshes composed of triangular faces. Small isolated triangular faces are identified and deleted to obtain a three-dimensional model of roadbed construction.

[0010] Optionally, a quadratic error measure grid simplification algorithm is used to perform lightweight processing on the roadbed construction three-dimensional model, including: Calculate the normal vector of each triangular face in the three-dimensional model of roadbed construction, calculate the sum of the differences between the normal vector of the triangular face and the adjacent normal vectors as the curvature of the triangular face, the adjacent normal vector of the triangular face is the normal vector of the triangular face connected to the triangular face; The error weight of the triangular face is calculated based on the curvature, and the plane weighted error matrix of the triangular face is calculated based on the error weight and the normal vector. The plane weighted error matrix is ​​used to measure the distance from the vertex of the triangular face to the plane. The calculation formula of the error weight is: ; in: represents the error weight corresponding to the curvature C, represents the preset error weight threshold; represents the mean curvature of the adjacent triangles corresponding to the curvature C, and F represents the normal vector of the triangle corresponding to the curvature C. represents the gravity vector, ; The accumulated weighted error matrices of all adjacent planes of the triangular vertex are calculated as the error matrix of the triangular vertex; Select triangle vertices and triangle vertices with shared edges in the order of the error matrix L2 norm value from low to high to form triangle vertex pairs, calculate the new triangle vertex with the smallest error matrix after merging the triangle vertex pairs, delete the merged triangle vertex, and update the adjacent triangle to point to the new triangle vertex; Vertices are repeatedly merged, the error matrix is ​​updated, and the number of triangles is gradually reduced until the set total number of triangles is reached. The merged 3D model is processed using Laplacian smoothing as a 3D model for lightweight roadbed construction.

[0011] Optionally, the color vector in the roadbed fusion data is extracted, and the color vector is mapped to the corresponding position of the lightweight roadbed construction three-dimensional model to obtain the color-rendered lightweight roadbed construction three-dimensional model, and multi-source roadbed data is collected in real time during the roadbed construction operation to dynamically update the lightweight roadbed construction three-dimensional model.

[0012] In order to solve the above problems, the present invention provides a three-dimensional visualization operation control system for intelligent construction of roadbed, which includes a server and a data acquisition device, and the server includes a data processing module and a three-dimensional visualization module: The data processing module is used to perform position correction on the multi-source roadbed data using an improved SLAM algorithm, and to perform multi-source fusion on the multi-source roadbed data after position correction to obtain roadbed fusion data; The three-dimensional visualization module is used to construct a three-dimensional model of roadbed construction based on the roadbed fusion data, use a quadratic error measure grid simplification algorithm to lightweight the three-dimensional model of roadbed construction to obtain a lightweight three-dimensional model of roadbed construction, perform dynamic operation rendering on the lightweight three-dimensional model of roadbed construction, and perform visual operation rendering on the roadbed construction process; The data acquisition device is used to collect multi-source roadbed data.

[0013] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising: A memory storing at least one instruction; Communication interface, enabling electronic equipment to communicate; and The processor executes the instructions stored in the memory to implement the above-mentioned three-dimensional visualization operation control method for intelligent construction of roadbed.

[0014] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned three-dimensional visualization operation control method for intelligent roadbed construction.

[0015] Compared with the prior art, the present invention proposes a three-dimensional visualization operation control method for intelligent roadbed construction, which has the following advantages: First, this scheme proposes a global position correction method, which uses a position matching algorithm to correct the position of multi-source roadbed data. By taking the data acquisition channel as a factor, the position deviation between different factors is identified, and a multi-source factor error function is constructed. The multi-source factor error function is solved to obtain the correction result of the global position matching, and the quality weight that characterizes the quality of the factor is introduced in the solution process. The higher the quality weight, the higher the position accuracy of the collected data. Finally, high-precision, aligned multi-source fused roadbed data is generated. According to the multi-dimensional feature entropy of different roadbed areas, the roadbed areas with complex terrain are screened for multiple resampling to improve the accuracy of three-dimensional modeling in complex roadbed areas.

[0016] At the same time, this scheme proposes a three-dimensional modeling method, which calculates the filtering scale information of the multi-source fusion roadbed data based on multi-source information. If the filtering scale information is higher than the preset filtering threshold, it means that the multi-source fusion roadbed data is noise data. An octree is constructed to store the multi-source fusion roadbed data in the filtered roadbed fusion data, and the filtered roadbed fusion data is spatially hierarchical. The Poisson equation of different spatial levels is constructed, and the Poisson equation is solved to obtain the smooth implicit surface equation of different roadbed areas. The isosurface is extracted to generate a closed triangular mesh, and the curvature is used to identify key areas such as slopes and faults. The high-curvature triangular faces are key areas such as slopes and faults, and the low-curvature triangular faces are flat areas. By introducing the local curvature mean, the curvature is normalized to ensure that the curvature distribution is consistent on different triangular faces. The triangular face vertices in the low-curvature area are merged based on the error matrix to reduce the complexity of the triangular mesh, and the merging of the triangular vertices in the high-curvature area is limited to ensure the integrity of features such as slopes and faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of a process flow of a three-dimensional visualization operation control method for intelligent construction of roadbed provided by an embodiment of the present invention; Figure 2 A functional module diagram of a three-dimensional visualization operation control system for intelligent roadbed construction provided by an embodiment of the present invention; Figure 2 In: 100 three-dimensional visualization operation control system for intelligent roadbed construction, 101 data processing module, 102 three-dimensional visualization module, 103 data acquisition device; The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0018] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0019] The embodiment of the present application provides a three-dimensional visualization operation control method for intelligent construction of roadbed. The execution subject of the three-dimensional visualization operation control method for intelligent construction of roadbed includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the three-dimensional visualization operation control method for intelligent construction of roadbed can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0020] Reference Figure 1 , Embodiment 1 of the present invention is: S1: Collect multi-source roadbed data, use the improved SLAM algorithm to perform position correction on the multi-source roadbed data, and obtain the multi-source roadbed data after position correction.

[0021] Using a laser radar to transmit a laser radar signal to the roadbed area, constructing three-dimensional point cloud data of the roadbed area based on the laser radar echo signal, using a ground-based device to collect radar echo data of the roadbed area, the ground-based device transmits a coherent radar signal to the roadbed area, extracts phase information and coherence in the echo signal to form the radar echo data, uses a global navigation satellite system to obtain satellite positioning data of the roadbed area, and forms multi-source roadbed data with the three-dimensional point cloud data of the roadbed area, the radar echo data and the satellite positioning data; The three-dimensional point cloud data includes three-dimensional position coordinates and color vectors; The radar echo data includes three-dimensional position coordinates, signal amplitude, phase information, interference phase and coherence; The satellite positioning data includes three-dimensional position coordinates and movement speed; The multi-source roadbed data is represented in the form of: ; ; in: G represents multi-source roadbed data, Represents a 3D point cloud data set, represents a radar echo data set, represents the satellite positioning data set, and N represents the number of data collection; Represents a 3D point cloud data set The nth group of 3D point cloud data in Represents a radar echo data set The nth group of radar echo data in Represents a collection of satellite positioning data The nth group of satellite positioning data in; As an embodiment of the present invention, the ground-based equipment is a ground-based InSAR (GB-InSAR, Ground-Based Interferometric Synthetic Aperture Radar), which uses coherent radar signals to perform high-precision measurements of surface deformation, foundation settlement, landslide monitoring, etc.; Specifically, the ground-based InSAR continuously transmits two radar signals to the same three-dimensional position coordinates of the roadbed area as coherent radar signals, extracts the phase information difference and phase information similarity of the two radar signals as the interference phase and coherence corresponding to the three-dimensional position coordinates, respectively, and extracts the signal negative value and phase information of the second radar signal to form radar echo data; The improved SLAM algorithm is used to construct a multi-source factor error function of the multi-source roadbed data, wherein the multi-source factor error function takes the global position correction parameter as a variable and the degree of matching of the three-dimensional position coordinates of the multi-source roadbed data after the global position correction as a function value. The improved SLAM algorithm is used to perform position matching on the multi-source roadbed data, and the three-dimensional position coordinates in the three-dimensional point cloud data, radar echo data and satellite positioning data are used as nodes, and the multi-source factor error functions between the nodes are used as edges to construct a factor graph.

[0022] The expression of the multi-source factor error function is: ; ; ; ; ; ; in: Indicates The quality weight of the error factor, represents the global position correction parameter to be solved, It represents the position correction parameters of 3D point cloud data, radar echo data and satellite positioning data respectively. Indicates The first to third factor errors are the factor errors of three-dimensional point cloud data, radar echo data and satellite positioning data, respectively. The first to third factor error functions are the factor error functions of three-dimensional point cloud data, radar echo data and satellite positioning data, respectively. express The standard deviation of the N three-dimensional position coordinates in ; represents the L1 norm, Represents 3D point cloud data , radar echo data And satellite positioning data The three-dimensional position coordinates in ; Representing a collection Zhongyu The nearest three-dimensional position coordinates; Representing a collection Zhongyu The nearest three-dimensional position coordinates; Representing a collection Zhongyu The nearest 3D position coordinates.

[0023] With the goal of minimizing the multi-source factor error function, the multi-source factor error function is solved to obtain global position correction parameters, including: Initializing and generating global position correction parameters, and using the generated global position correction parameters as input values ​​of the multi-source factor error function; Calculate the first-order gradient of the multi-source factor error function as the Jacobian matrix of the generated global position correction parameters, and calculate the second-order inverse of the Jacobian matrix as the Hessian matrix of the generated global position correction parameters; Iterating the generated global position correction parameters based on the Jacobian matrix and the Hessian matrix, and re-iterating the Jacobian matrix and the Hessian matrix of the global position correction parameters; As an embodiment of the present invention, the iterative formula of the global position correction parameter is: ; in: represents the tth iteration result of the global position correction parameter, express The Jacobian matrix of express The Hessian matrix of Indicates the preset maximum correction parameter iteration result; After each iteration, the rate of change of the global position correction parameter is calculated. If the rate of change is less than a preset threshold, the iteration is terminated and the current iteration result is used as the global position correction parameter obtained by the solution. Specifically, the global position correction parameter The rate of change is: ; The three-dimensional position coordinates in the multi-source roadbed data are corrected using the global position correction parameters to obtain multi-source roadbed data after position correction.

[0024] As a preferred embodiment of the present invention, the three-dimensional position coordinates The correction formula is: ; ; ; in: These are the position correction parameters of the solved three-dimensional point cloud data, radar echo data, and satellite positioning data; The three-dimensional position coordinates are The correction result of .

[0025] S2: Perform multi-source fusion on the multi-source roadbed data after position correction to obtain roadbed fusion data.

[0026] The multi-source roadbed data after position correction is subjected to multi-source fusion, including: Calculating the three-dimensional position coordinate distances among the three-dimensional point cloud data, radar echo data, and satellite positioning data in the multi-source roadbed data after position correction, and fusing the data from different sources whose three-dimensional position coordinate distances are less than a preset distance threshold as multi-source fused roadbed data under the same three-dimensional position coordinates; Based on the multi-source fusion roadbed data, the multi-dimensional characteristic entropy of different roadbed areas is calculated, where the larger the multi-dimensional characteristic entropy, the more complex the surface terrain. In the roadbed area where the multi-dimensional characteristic entropy is higher than the preset entropy threshold, multiple multi-source roadbed data collection is performed. Based on the global position correction parameters, the three-dimensional position coordinates in the newly collected multi-source roadbed data are corrected and integrated into the multi-source fusion roadbed data to form roadbed fusion data. The calculation process of the multidimensional feature entropy is: Obtain multi-source fused roadbed data sets of different roadbed areas, and calculate the multi-dimensional feature entropy of the roadbed area. The multi-source fused roadbed data set of the rth roadbed area is , , R represents the total number of roadbed areas, Indicates the e-th group of multi-source fused roadbed data whose three-dimensional position coordinates are located in the r-th roadbed area, Indicates the number of multi-source fused roadbed data with three-dimensional position coordinates located in the r-th roadbed area; ; in: In turn, it represents multi-source fusion roadbed data The three-dimensional position coordinates, color vector, signal amplitude, phase information, interference phase, coherence and movement speed; The multidimensional characteristic entropy of the r-th roadbed area is: ; ; ; ; in: represents the multidimensional characteristic entropy of the r-th roadbed area, represents the point cloud entropy of the r-th roadbed area, represents the echo entropy of the r-th roadbed area, represents the velocity entropy of the r-th roadbed area, In turn, they represent the preset point cloud entropy extreme value, echo entropy extreme value, and velocity entropy extreme value; represents the mean coherence value in the r-th roadbed area, represents the mean speed in the rth roadbed area, represents the standard deviation of speed in the rth roadbed area; Specifically, the roadbed fusion data is represented as follows: ; in: represents the roadbed fusion data, represents the mth group of multi-source fusion roadbed data, M represents the number of groups of multi-source fusion roadbed data in the roadbed fusion data g, Sequentially indicate The three-dimensional position coordinates, color vector, signal amplitude, phase information, interference phase, coherence and movement speed in is the fusion result of the mth group of data from different sources whose three-dimensional position coordinate distance is less than the preset distance threshold. The 3D position coordinates in are the average of the fused 3D point cloud data, radar echo data, and satellite positioning data.

[0027] S3: Construct a 3D model of roadbed construction based on roadbed fusion data.

[0028] The three-dimensional model of roadbed construction includes a filtering module and a three-dimensional surface reconstruction module. The filtering module is used to extract multi-source information of each group of multi-source fused roadbed data in the roadbed fusion data, and perform multi-scale filtering on the multi-source fused roadbed data to obtain filtered roadbed fusion data. The three-dimensional surface reconstruction module is used to extract three-dimensional position coordinates and multi-source information from the filtered roadbed fusion data, and perform three-dimensional network reconstruction using a Poisson reconstruction method to obtain a three-dimensional model of roadbed construction. The multi-source information in the multi-source fusion roadbed data is the signal amplitude, phase information, interference phase, coherence and movement speed in the multi-source fusion roadbed data; The multi-scale filtering processing method is as follows: based on multi-source information, filter scale information of multi-source fusion roadbed data is calculated. If the filter scale information is higher than a preset filter threshold, it indicates that the multi-source fusion roadbed data is noise data; the noise data in the roadbed fusion data is filtered and cleaned to obtain filtered roadbed fusion data; specifically, the multi-source fusion roadbed data The filtering scale information is : ; in: The preset signal amplitude extreme value, phase information extreme value, interference phase extreme value, coherence extreme value and motion speed extreme value; The Poisson reconstruction method is: using principal component analysis to calculate the normal vector of each group of multi-source fused roadbed data in the filtered roadbed fusion data, wherein the higher the coherence, the higher the weight of the multi-source fused roadbed data in the normal vector calculation process; An octree is constructed to store the multi-source fused roadbed data in the filtered roadbed fusion data, and the filtered roadbed fusion data is spatially hierarchically divided; specifically, the scheme adopts adaptive resolution to perform spatial hierarchical division, and for a multi-source fused roadbed data set with a strong signal amplitude and a high coherence, a finer resolution is used, and for other multi-source fused roadbed data, a lower resolution is used; Poisson equations at different spatial levels are constructed, the Poisson equations are solved, smooth implicit surface equations of different roadbed areas are obtained, isosurfaces are extracted, a closed triangular mesh is generated, the triangular mesh is composed of triangular faces, small isolated triangular faces are identified and deleted, and a three-dimensional model of roadbed construction is obtained. As an embodiment of the present invention, more details are retained in the spatial level with high coherence to obtain a smooth implicit surface equation with a higher degree of complexity, and in the spatial level with low coherence, the smooth implicit surface equation is appropriately smoothed to reduce the influence of noise on the three-dimensional modeling process.

[0029] S4: The quadratic error measure grid simplification algorithm is used to lightweight the roadbed construction three-dimensional model to obtain a lightweight roadbed construction three-dimensional model.

[0030] The three-dimensional model of the roadbed construction is lightweighted by using a quadratic error measure mesh simplification algorithm, including: Calculate the normal vector of each triangular face in the three-dimensional model of roadbed construction, calculate the sum of the differences between the normal vector of the triangular face and the adjacent normal vectors as the curvature of the triangular face, the adjacent normal vector of the triangular face is the normal vector of the triangular face connected to the triangular face; The error weight of the triangular face is calculated based on the curvature, and the plane weighted error matrix of the triangular face is calculated based on the error weight and the normal vector. The plane weighted error matrix is ​​used to measure the distance from the vertex of the triangular face to the plane. The calculation formula of the error weight is: ; in: represents the error weight corresponding to the curvature C, represents the preset error weight threshold; represents the mean curvature of the adjacent triangles corresponding to the curvature C, and F represents the normal vector of the triangle corresponding to the curvature C. represents the gravity vector, ; The accumulated weighted error matrices of all adjacent planes of the triangular vertex are calculated as the error matrix of the triangular vertex; Select triangle vertices and triangle vertices with shared edges in the order of the error matrix L2 norm value from low to high to form triangle vertex pairs, calculate the new triangle vertex with the smallest error matrix after merging the triangle vertex pairs, delete the merged triangle vertex, and update the adjacent triangle to point to the new triangle vertex; Merge vertices repeatedly, update the error matrix, and gradually reduce the number of triangular faces until the set total number of triangular faces is reached. Use Laplacian smoothing to process the merged three-dimensional model as a three-dimensional model for lightweight roadbed construction. As a preferred embodiment of the present invention, curvature is used to identify key areas such as slopes and faults, where high-curvature triangular faces are key areas such as slopes and faults, and low-curvature triangular faces are flat areas. By introducing the local curvature mean, the curvature is normalized to ensure that the curvature distribution is consistent on different triangular faces. The triangular face vertices in the low-curvature area are merged based on the error matrix to reduce the complexity of the triangular mesh, and the merging of triangular vertices in the high-curvature area is limited to ensure that the features such as slopes and faults are complete.

[0031] S5: Dynamically render the 3D model of lightweight roadbed construction and visualize the roadbed construction process.

[0032] The color vector in the roadbed fusion data is extracted, and the color vector is mapped to the corresponding position of the lightweight roadbed construction three-dimensional model to obtain the color-rendered lightweight roadbed construction three-dimensional model. In addition, multi-source roadbed data is collected in real time during the roadbed construction operation to dynamically update the lightweight roadbed construction three-dimensional model.

[0033] Embodiment 2: like Figure 2As shown, it is a functional module diagram of a three-dimensional visualization operation control system 100 for intelligent roadbed construction provided in one embodiment of the present invention, which can implement the three-dimensional visualization operation control method for intelligent roadbed construction in Example 1.

[0034] According to the functions to be implemented, the roadbed intelligent construction 3D visualization operation control system 100 may include a data processing module 101, a 3D visualization module 102 and a data acquisition device 103. The module described in the present invention may also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions.

[0035] The data processing module 101 is used to perform position correction on multi-source roadbed data using an improved SLAM algorithm, and to perform multi-source fusion on the multi-source roadbed data after position correction to obtain roadbed fusion data; The three-dimensional visualization module 102 is used to construct a three-dimensional model of roadbed construction based on the roadbed fusion data, use a quadratic error measure grid simplification algorithm to lightweight the three-dimensional model of roadbed construction to obtain a lightweight three-dimensional model of roadbed construction, perform dynamic operation rendering on the lightweight three-dimensional model of roadbed construction, and perform visual operation rendering on the roadbed construction process; The data acquisition device 103 is used to acquire multi-source roadbed data.

[0036] In detail, the modules in the roadbed intelligent construction three-dimensional visualization operation control system 100 in the embodiment of the present invention are used in the same manner as above. Figure 1 The technical means are the same as the three-dimensional visualization operation control method for intelligent roadbed construction described in, and can produce the same technical effects, so they will not be repeated here.

[0037] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0038] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof in this article are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0039] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0040] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A three-dimensional visualization operation control method for intelligent roadbed construction, characterized in that: The method comprises: S1: Collect multi-source roadbed data, use the improved SLAM algorithm to perform position correction on the multi-source roadbed data, and obtain the multi-source roadbed data after position correction; The multi-source roadbed data includes three-dimensional point cloud data, radar echo data and satellite positioning data of the roadbed area; S2: Perform multi-source fusion on the multi-source roadbed data after position correction to obtain roadbed fusion data; S3: Construct a 3D model of roadbed construction based on roadbed fusion data; S4: Use the quadratic error measure grid simplification algorithm to lightweight the roadbed construction three-dimensional model to obtain a lightweight roadbed construction three-dimensional model; S5: Dynamically render the 3D model of lightweight roadbed construction and visualize the roadbed construction process.

2. A three-dimensional visualization operation control method for intelligent roadbed construction according to claim 1, characterized in that: The three-dimensional point cloud data includes three-dimensional position coordinates and color vectors; The radar echo data includes three-dimensional position coordinates, signal amplitude, phase information, interference phase and coherence; The satellite positioning data includes three-dimensional position coordinates and movement speed; The multi-source roadbed data is represented in the form of: ； ； in: G represents multi-source roadbed data, Represents a 3D point cloud data set, represents a radar echo data set, represents the satellite positioning data set, and N represents the number of data collection; Represents a 3D point cloud data set The nth group of 3D point cloud data in Represents a radar echo data set The nth group of radar echo data in Represents a collection of satellite positioning data The nth group of satellite positioning data in; An improved SLAM algorithm is used to construct a multi-source factor error function of the multi-source roadbed data. The multi-source factor error function takes a global position correction parameter as a variable and takes the degree of matching of the three-dimensional position coordinates of the multi-source roadbed data after global position correction as a function value.

3. A three-dimensional visualization operation control method for intelligent roadbed construction according to claim 2, characterized in that: The expression of the multi-source factor error function is: ； ； ； ； ； ； in: Indicates The quality weight of the error factor, represents the global position correction parameter to be solved, It represents the position correction parameters of 3D point cloud data, radar echo data and satellite positioning data respectively. Indicates The first to third factor errors are the factor errors of three-dimensional point cloud data, radar echo data and satellite positioning data, respectively. The first to third factor error functions are the factor error functions of three-dimensional point cloud data, radar echo data and satellite positioning data, respectively. express The standard deviation of the N three-dimensional position coordinates in ; represents the L1 norm, Represents 3D point cloud data , radar echo data And satellite positioning data The three-dimensional position coordinates in ; Representing a collection Zhongyu The nearest three-dimensional position coordinates; Representing a collection Zhongyu The nearest three-dimensional position coordinates; Representing a collection Zhongyu The nearest 3D position coordinates.

4. A three-dimensional visualization operation control method for intelligent roadbed construction according to claim 3, characterized in that: With the goal of minimizing the multi-source factor error function, the multi-source factor error function is solved to obtain global position correction parameters, including: Initializing and generating global position correction parameters, and using the generated global position correction parameters as input values ​​of the multi-source factor error function; Calculate the first-order gradient of the multi-source factor error function as the Jacobian matrix of the generated global position correction parameters, and calculate the second-order inverse of the Jacobian matrix as the Hessian matrix of the generated global position correction parameters; Iterating the generated global position correction parameters based on the Jacobian matrix and the Hessian matrix, and re-iterating the Jacobian matrix and the Hessian matrix of the global position correction parameters; After each iteration, the rate of change of the global position correction parameter is calculated. If the rate of change is less than a preset threshold, the iteration is terminated and the current iteration result is used as the solved global position correction parameter; the three-dimensional position coordinates in the multi-source roadbed data are corrected using the global position correction parameter to obtain the multi-source roadbed data after position correction.

5. A three-dimensional visualization operation control method for intelligent roadbed construction according to claim 4, characterized in that: The multi-source roadbed data after position correction is subjected to multi-source fusion, including: Calculating the three-dimensional position coordinate distances among the three-dimensional point cloud data, radar echo data, and satellite positioning data in the multi-source roadbed data after position correction, and fusing the data from different sources whose three-dimensional position coordinate distances are less than a preset distance threshold as multi-source fused roadbed data under the same three-dimensional position coordinates; The multi-dimensional characteristic entropy of different roadbed areas is calculated based on the multi-source fused roadbed data. The larger the multi-dimensional characteristic entropy, the more complex the surface terrain. Multiple multi-source roadbed data collection is performed in the roadbed area where the multi-dimensional characteristic entropy is higher than the preset entropy threshold. Based on the global position correction parameters, the three-dimensional position coordinates in the newly collected multi-source roadbed data are corrected and integrated into the multi-source fused roadbed data to form roadbed fusion data.

6. A three-dimensional visualization operation control method for intelligent roadbed construction according to claim 1, characterized in that: The three-dimensional model of roadbed construction includes a filtering module and a three-dimensional surface reconstruction module. The filtering module is used to extract multi-source information of each group of multi-source fused roadbed data in the roadbed fusion data, and perform multi-scale filtering on the multi-source fused roadbed data to obtain filtered roadbed fusion data. The three-dimensional surface reconstruction module is used to extract three-dimensional position coordinates and multi-source information from the filtered roadbed fusion data, and perform three-dimensional network reconstruction using a Poisson reconstruction method to obtain a three-dimensional model of roadbed construction. The multi-source information in the multi-source fusion roadbed data is the signal amplitude, phase information, interference phase, coherence and movement speed in the multi-source fusion roadbed data; The multi-scale filtering processing method is as follows: based on multi-source information, filter scale information of multi-source fused roadbed data is calculated; if the filter scale information is higher than a preset filter threshold, it indicates that the multi-source fused roadbed data is noise data; the noise data in the roadbed fusion data is filtered and cleaned to obtain filtered roadbed fusion data; The Poisson reconstruction method is: using principal component analysis to calculate the normal vector of each group of multi-source fused roadbed data in the filtered roadbed fusion data, wherein the higher the coherence, the higher the weight of the multi-source fused roadbed data in the normal vector calculation process; Constructing an octree to store the multi-source fused roadbed data in the filtered roadbed fusion data, and performing spatial hierarchical division on the filtered roadbed fusion data; Poisson equations at different spatial levels are constructed and solved to obtain smooth implicit surface equations of different roadbed areas. Isosurfaces are extracted to generate closed triangular meshes composed of triangular faces. Small isolated triangular faces are identified and deleted to obtain a three-dimensional model of roadbed construction.

7. A three-dimensional visualization operation control method for intelligent roadbed construction according to claim 6, characterized in that: The three-dimensional model of the roadbed construction is lightweighted by using a quadratic error measure mesh simplification algorithm, including: Calculate the normal vector of each triangular face in the three-dimensional model of roadbed construction, and calculate the sum of the differences between the normal vector of the triangular face and the adjacent normal vectors as the curvature of the triangular face; The error weight of the triangular face is calculated based on the curvature, and the plane weighted error matrix of the triangular face is calculated based on the error weight and the normal vector. The plane weighted error matrix is ​​used to measure the distance from the vertex of the triangular face to the plane. The calculation formula of the error weight is: ； in: represents the error weight corresponding to the curvature C, represents the preset error weight threshold; represents the mean curvature of the adjacent triangles corresponding to the curvature C, and F represents the normal vector of the triangle corresponding to the curvature C. represents the gravity vector, ; The accumulated weighted error matrices of all adjacent planes of the triangular vertex are calculated as the error matrix of the triangular vertex; Select triangle vertices and triangle vertices with shared edges in the order of the error matrix L2 norm value from low to high to form triangle vertex pairs, calculate the new triangle vertex with the smallest error matrix after merging the triangle vertex pairs, delete the merged triangle vertex, and update the adjacent triangle to point to the new triangle vertex; Vertices are repeatedly merged, the error matrix is ​​updated, and the number of triangles is gradually reduced until the set total number of triangles is reached. The merged 3D model is processed using Laplacian smoothing as a 3D model for lightweight roadbed construction.

8. A three-dimensional visualization operation control method for intelligent roadbed construction according to claim 7, characterized in that: The color vector in the roadbed fusion data is extracted, and the color vector is mapped to the corresponding position of the lightweight roadbed construction three-dimensional model to obtain the color-rendered lightweight roadbed construction three-dimensional model. In addition, multi-source roadbed data is collected in real time during the roadbed construction operation to dynamically update the lightweight roadbed construction three-dimensional model.

9. A three-dimensional visualization operation control system for intelligent roadbed construction, characterized in that: The roadbed intelligent construction three-dimensional visualization operation control system includes a server and a data acquisition device, and the server includes a data processing module and a three-dimensional visualization module: The data processing module is used to perform position correction on the multi-source roadbed data using an improved SLAM algorithm, and to perform multi-source fusion on the multi-source roadbed data after position correction to obtain roadbed fusion data; The three-dimensional visualization module is used to construct a three-dimensional model of roadbed construction based on the roadbed fusion data, use a quadratic error measure grid simplification algorithm to lightweight the three-dimensional model of roadbed construction to obtain a lightweight three-dimensional model of roadbed construction, perform dynamic operation rendering on the lightweight three-dimensional model of roadbed construction, and perform visual operation rendering on the roadbed construction process; The data acquisition device is used to collect multi-source roadbed data; To realize a three-dimensional visualization operation control method for intelligent roadbed construction as described in any one of claims 1-8.

Images