Highway engineering construction quality global rapid sensing method and system

Abstract

The invention relates to the technical field of highway engineering construction quality detection, in particular to a global rapid sensing method and system for highway engineering construction quality. The method comprises the following steps: constructing a BIM design model of highway engineering, and sampling the BIM design model to form a BIM point cloud data set; the method comprises the following steps: acquiring LiDAR point cloud data of a construction area, and performing data cleaning on the LiDAR point cloud data to obtain effective LiDAR point cloud data; initial space alignment and self-adaptive fine registration are carried out on the BIM point cloud data set and the effective LiDAR point cloud data; and performing curved surface fitting on the BIM point cloud data set and the effective LiDAR point cloud data by adopting a least square method, and evaluating the construction quality of the highway engineering in combination with a registration result and a curved surface fitting result. By comparing the BIM design model of the highway engineering with the actual point cloud data difference, the problem of quickly quantifying the construction quality is solved, and real-time and reliable data support is provided for construction management.

Description

Technical Field

[0001] The present invention relates to the technical field of highway engineering construction quality detection, and in particular to a method and system for rapid global perception of highway engineering construction quality. Background Art

[0002] Traditional highway construction quality inspections mainly rely on manual sampling, single-point sensor monitoring, or drone inspections of local areas. However, manual sampling is inefficient, has insufficient coverage, and is greatly affected by subjective factors. Single-point sensors can only obtain local data and cannot reflect the overall construction quality.

[0003] With the rapid development of information technology, the application of BIM (Building Information Modeling) technology and LiDAR (LiDAR) point cloud technology in the field of engineering construction has shown a booming trend. With its powerful three-dimensional modeling capabilities and rich attribute information integration functions, BIM technology provides a full life cycle information management platform for engineering projects from design, construction to operation and maintenance, greatly improving the level of refinement of engineering management. Although the BIM model can accurately describe the design intent, it lacks the ability to perceive the real-time dynamic changes of the construction site, and it is difficult to reflect the deviation between the actual state and the design model caused by various factors during the construction process.

[0004] LiDAR point cloud technology, with its efficient and accurate three-dimensional data collection capabilities, provides a new technical means for real-time monitoring of construction sites. Through the LiDAR equipment carried by drones, high-precision three-dimensional point cloud data of the construction site can be quickly obtained to intuitively display the topography, building outlines and construction progress. Although LiDAR point cloud data can accurately reflect the spatial form of the construction site, it lacks the ability to associate the semantic information and attribute data of the construction object, and it is difficult to be directly used for quantitative analysis and deviation comparison of construction quality and progress. Therefore, how to efficiently combine the two to achieve global rapid perception and deviation comparison of highway construction quality and progress has become a key technical problem that needs to be solved in the current engineering construction field. Summary of the invention

[0005] In view of the deficiencies of existing methods and the requirements of practical applications, in order to solve the problem of quickly quantifying construction quality and provide real-time and reliable data support for construction management. On the one hand, the present invention provides a method for quickly perceiving the overall construction quality of highway engineering, including the following steps: constructing a BIM design model of highway engineering, sampling the BIM design model to form a BIM point cloud data set; obtaining LiDAR point cloud data of the construction area, cleaning the LiDAR point cloud data to obtain effective LiDAR point cloud data; performing initial spatial alignment and adaptive fine registration on the BIM point cloud data set and the effective LiDAR point cloud data; using the least squares method to perform surface fitting on the BIM point cloud data set and the effective LiDAR point cloud data, and evaluating the construction quality of highway engineering by combining the registration results and the surface fitting results.

[0006] The present invention solves the problem of quickly quantifying construction quality by fusing and comparing the BIM design model of highway engineering and the actual point cloud data, and provides real-time and reliable data support for construction management.

[0007] Optionally, the sampling of the BIM design model to form a BIM point cloud data set includes the following steps: Converting the BIM design model into a triangular mesh and uniformly sampling from the triangular faces; based on the sampling results, generating BIM point clouds according to the centroids of the triangles. The present invention converts the BIM design model into BIM point clouds, which is beneficial to fusing and comparing the BIM design model and the actual point cloud data.

[0008] Optionally, the data cleaning of the LiDAR point cloud data to obtain effective LiDAR point cloud data includes the following steps: performing erosion and dilation operations on the LiDAR point cloud data; setting an elevation difference threshold model, and performing opening operations on the results of the erosion and dilation operations according to the elevation difference threshold to obtain the effective LiDAR point cloud data. By removing outliers and noise, the present invention is beneficial to more accurately identifying and removing invalid data points, retaining more effective point cloud data, and further improving the perception accuracy of the present invention.

[0009] Optionally, the initial spatial alignment of the BIM point cloud data set and the effective LiDAR point cloud data includes the following steps: dividing the BIM point cloud data set and the effective LiDAR point cloud data according to elevation features and point cloud density differences; constructing an adaptive threshold model, and extracting a feature point set from the divided BIM point cloud data set and effective LiDAR point cloud data through the adaptive threshold model; using the feature point set to calculate a rigid body transformation matrix to complete the initial spatial alignment. By performing initial spatial alignment, the present invention reduces the computational amount of point cloud fine registration, and is further beneficial to improving the perception efficiency of the present invention.

[0010] Optionally, the adaptive threshold model satisfies the following formula: , where represents the adaptive threshold, represents the mean of the local curvatures of all points in the point cloud, represents the adjustment coefficient, which can be adjusted according to the actual situation of the point cloud data, represents the standard deviation of the local curvature values of all points in the point cloud. By dynamically adjusting the threshold, the present invention adapts to the local changes of different point clouds, effectively screens out key points, and improves the registration accuracy.

[0011] Optionally, the adaptive fine registration of the BIM point cloud dataset and the effective LiDAR point cloud data includes the following steps: Construct a fine registration error model and a fine registration objective function; use the fine registration error model to adjust the weight coefficients in the fine registration objective function, and obtain an accurate rigid body transformation matrix through the adjusted fine registration objective function to complete the adaptive fine registration. By introducing geometric features such as bridge bearings and tunnel linings as constraint conditions, the present invention further optimizes the registration accuracy.

[0012] Optionally, the fine registration error model satisfies the following formula: , where represents the registration error, represents the number of point clouds participating in the registration, represents the rotation matrix of i.e., the coordinate rotation from the BIM point cloud to the LiDAR point cloud, represents the three-dimensional coordinate of the th point in the BIM point cloud, represents the translation vector of i.e., the coordinate translation from the BIM point cloud to the LiDAR point cloud,

[0013] represents the three-dimensional coordinate of the , where represents the value of the fine registration objective function, represents the rotation matrix of represents The translation vector represents the coordinate translation from the BIM point cloud to the LiDAR point cloud. represents the total number of BIM points. represents the total number of LiDAR points. represents the BIM point and the LiDAR point between the weight coefficients. represents the th three-dimensional point coordinate in the BIM point cloud. represents the th three-dimensional point coordinate in the LiDAR point cloud.

[0014] Optionally, evaluating the construction quality of the highway project by combining the registration result and the surface fitting result includes the following steps: Based on the point cloud data on the fitted surface, calculate the first Euclidean distance between the surface points of each component; generate a deviation data set according to the first Euclidean distance, and use the deviation data set to evaluate the construction quality of the highway project. The present invention uses the registered point cloud data to fit the surface, and then generates deviation data according to the distances of each component in the surface, which is beneficial to clearly and objectively evaluate the construction quality of the highway project.

[0015] In a second aspect, in order to efficiently execute a method for rapid global perception of the construction quality of a highway project provided by the present invention, the present invention further provides a system for rapid global perception of the construction quality of a highway project, including a processor, an input device, an output device, and a memory. The processor, the input device, the output device, and the memory are interconnected. Among them, the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute a method for rapid global perception of the construction quality of a highway project as described in the first aspect of the present invention. The system for rapid global perception of the construction quality of a highway project according to the present invention has a compact structure and stable performance, and can stably execute a method for rapid global perception of the construction quality of a highway project provided by the present invention, further improving the overall applicability and practical application ability of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flowchart of a method for rapid global perception of the construction quality of a highway project provided by an embodiment of the present invention; Figure 2 is a framework diagram of a system for rapid global perception of the construction quality of a highway project provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] Specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described here are only for illustrative purposes and are not intended to limit the present invention. In the following description, in order to provide a thorough understanding of the present invention, a large number of specific details are set forth. However, it will be apparent to those of ordinary skill in the art that the present invention may be practiced without these specific details. In other instances, well-known circuits, software, or methods have not been specifically described in order to avoid obscuring the present invention.

[0018] Throughout the specification, references to "one embodiment", "an embodiment", "one example", or "an example" mean that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment of the present invention. Thus, the phrases "in one embodiment", "in an embodiment", "one example", or "an example" appearing throughout the specification are not necessarily all referring to the same embodiment or example. Additionally, the specific features, structures, or characteristics may be combined in any suitable combination and / or sub-combination in one or more embodiments or examples. Further, those of ordinary skill in the art should understand that the diagrams provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0019] Please refer to Figure 1 , to solve the problem of quickly quantifying construction quality and provide real-time and reliable data support for construction management. The present invention provides a method for rapid global perception of highway engineering construction quality. As Figure 1 shown, in one embodiment, the method includes the following steps: S1. Construct a BIM design model of the highway project, sample the BIM design model, and form a BIM point cloud dataset.

[0020] In an embodiment, use highway engineering BIM design software to construct a three-dimensional BIM model including components such as subgrades, bridges, and tunnels, and associate the semantic information of the components, including information such as station numbers, design elevations, and material properties, to form a BIM design model with complete attribute information.

[0021] Further, the sampling of the BIM design model to form a BIM point cloud dataset includes the following steps: S11. Convert the BIM design model into a triangular mesh and uniformly sample from the triangular faces.

[0022] Specifically, use BIM software to convert the BIM design model into a triangular mesh, and then uniformly sample from the triangular faces.

[0023] S12. Based on the sampling result, generate a BIM point cloud according to the centroid of the triangle.

[0024] In the embodiment, based on the adoption result, a BIM point cloud is generated according to the centroid of a triangle, which satisfies the following formula: , where represents the BIM point cloud coordinates, , represents the centroid coordinate weight coefficient inside the triangle, represents the weight close to vertex , represents the weight close to vertex , , , represent the three vertex coordinates of the triangle.

[0025] S2. Obtain the LiDAR point cloud data of the construction area, and perform data cleaning on the LiDAR point cloud data to obtain effective LiDAR point cloud data.

[0026] Use a drone to carry a LiDAR device to perform three-dimensional scanning on the construction site according to a preset route to obtain high-precision point cloud data, and the coverage range includes construction areas such as subgrade, bridge, and tunnel.

[0027] Furthermore, the step of performing data cleaning on the LiDAR point cloud data to obtain effective LiDAR point cloud data includes the following steps: S21. Perform erosion and dilation operations on the LiDAR point cloud data.

[0028] Design dilation operators and erosion operators for the data points in the LiDAR point cloud data, and their definition methods are as follows: , , where represents the structural element window. The result of the dilation operation is the height value of the highest point within the structural element where the point is located, and the result of the erosion operation is the height value of the corresponding lowest point.

[0029] S22. Set an elevation difference threshold model, and perform an opening operation on the result of the erosion and dilation operations according to the elevation difference threshold to obtain the effective LiDAR point cloud data.

[0030] Specifically, the elevation difference threshold model satisfies the following formula: , where represents the elevation difference threshold during the k-th filtering, represents the initial elevation difference, represents the size of the structural element window during the k-th filtering, represents the slope parameter, represents the cell size, Represents the maximum elevation difference. The threshold is dynamically set according to the elevation difference threshold model, which can effectively ensure that all objects are recognized during the opening operation.

[0031] Further, perform an opening operation on the erosion and dilation operation result according to the elevation difference threshold to obtain the effective LiDAR point cloud data, which satisfies the following formula: , where k represents the current filtering times, k = 1, 2,..., M, M represents the maximum filtering times, b represents the base of the exponential equation, used to control the growth speed and trend with the increase of k, and is valued according to the actual situation. The maximum structural element window size value is .

[0032] S3. Perform initial spatial alignment and adaptive fine registration on the BIM point cloud dataset and the effective LiDAR point cloud data.

[0033] In the embodiment, performing initial spatial alignment on the BIM point cloud dataset and the effective LiDAR point cloud data includes the following steps: S31. Segment the BIM point cloud dataset and the effective LiDAR point cloud data according to the elevation feature and the point cloud density difference.

[0034] Specifically, based on the elevation feature and the point cloud density difference, use the region growing algorithm to segment the point cloud into terrain point cloud and structure point cloud, forming a layered point cloud dataset. The terrain point cloud region is used as the terrain point cloud dataset, and the structure point cloud region is used as the structure point cloud dataset. It can be understood that segmenting the point cloud data and performing subsequent registration steps on each segmented part can effectively improve the registration accuracy and efficiency.

[0035] S32. Construct an adaptive threshold model, and extract a feature point set from the segmented BIM point cloud dataset and effective LiDAR point cloud data through the adaptive threshold model.

[0036] For the terrain point cloud, pay attention to the undulation change of the terrain. For the structure point cloud, pay attention to the geometric features of the structure. Extract linear features such as road centerlines and stake number marks from the BIM point cloud and LiDAR point cloud to generate a feature point set.

[0037] Specifically, obtaining the feature point set from the BIM point cloud means extracting linear features such as geometric definitions, axes, stake numbers, and edge lines of structures from the geometric data of the BIM model; obtaining the feature point set from the LiDAR point cloud is based on the above-mentioned point cloud dataset segmentation, selecting the terrain point cloud, and using the RANSAC algorithm to extract linear features.

[0038] Furthermore, when extracting features and judging the characteristics of points, the neighborhood point set provides local geometric information. By analyzing the point distribution in the neighborhood, the identification of feature points can be adjusted or confirmed, avoiding misidentification and improving the accuracy of registration.

[0039] Specifically, the neighborhood point set is determined by the k-nearest neighbor algorithm, and then by dynamically adjusting the threshold to adapt to the local changes of different point clouds, key points are effectively screened out to form a feature point set.

[0040] Furthermore, the adaptive threshold model satisfies the following formula: , where, represents the adaptive threshold, represents the mean of the local curvatures of all points in the point cloud, represents the adjustment coefficient, which can be adjusted according to the actual situation of the point cloud data, represents the standard deviation of the local curvature values of all points in the point cloud.

[0041] S33. Calculate the rigid body transformation matrix using the feature point set to complete the initial spatial alignment.

[0042] Specifically, the random sample consensus algorithm and the feature matching of the graph neural network are used to calculate the rigid body transformation matrix of the feature point set, realizing the initial spatial alignment of the BIM point cloud and the LiDAR point cloud.

[0043] Furthermore, when performing feature matching, the BIM point cloud feature point set and the LiDAR cloud feature point set the cosine similarity between the feature vectors of the feature points in the point. The cosine similarity satisfies the following formula: , represents the cosine similarity between the th feature point in the point cloud feature point set and the th feature point in the point cloud feature point set , represents the th feature point in the point cloud feature point set , represents the th feature point in the point cloud feature point set , represents the feature vector of the th feature point in the point cloud feature point set , represents the feature vector of the th feature point in the point cloud feature point set .

[0044] In yet another embodiment, the adaptive fine registration of the BIM point cloud dataset and the valid LiDAR point cloud data means using the ICP algorithm, identifying and extracting a point set of geometric features through the RANSAC algorithm, introducing geometric features such as bridge bearings and tunnel linings as constraint conditions, optimizing the registration accuracy, and generating an accurate rigid body transformation matrix by minimizing the Euclidean distance between the point clouds to achieve high-precision registration of the BIM point cloud and the LiDAR point cloud.

[0045] Specifically, it includes the following steps: S34. Construct a fine registration error model and a fine registration objective function.

[0046] In the embodiment, the fine registration error model satisfies the following formula: , where, represents the registration error, represents the number of point clouds participating in the registration, represents the rotation matrix of i.e., the coordinate system rotation from the BIM point cloud to the LiDAR point cloud, represents the three-dimensional point coordinates of the th point in the BIM point cloud, represents the translation vector of i.e., the coordinate system translation from the BIM point cloud to the LiDAR point cloud,

[0047] Further, the fine registration objective function satisfies the following formula: , where, represents the value of the fine registration objective function, represents the rotation matrix of i.e., the coordinate system rotation from the BIM point cloud to the LiDAR point cloud, represents the translation vector of i.e., the coordinate system translation from the BIM point cloud to the LiDAR point cloud, represents the total number of BIM points, represents the total number of LiDAR points, represents the weight coefficient between the BIM point and the LiDAR point in the BIM point cloud, represents the three-dimensional point coordinates of the th point in the LiDAR point cloud.

[0048] S35. Use the refined registration error model to adjust the weight coefficients in the refined registration objective function, and obtain an accurate rigid body transformation matrix through the adjusted refined registration objective function to complete adaptive refined registration.

[0049] Specifically, set a preset threshold for the refined registration error according to industry specifications and the characteristics of the point cloud data. The preset threshold can be selected as 0.05 -0.2 . If the refined registration error increases or varies beyond the preset threshold, dynamically adjust the weight coefficients in the refined registration objective function.

[0050] Furthermore, dynamically adjust the weight coefficients in the refined registration objective function to satisfy the following formula:

[0051] where, represents the Euclidean distance between point pairs, represents the distance threshold parameter, .

[0052] Furthermore, obtain an accurate rigid body transformation matrix through the adjusted refined registration objective function to complete adaptive refined registration.

[0053] S4. Use the least squares method to perform surface fitting on the BIM point cloud dataset and the effective LiDAR point cloud data, and evaluate the construction quality of the highway project by combining the registration result and the surface fitting result.

[0054] In the embodiment, using the least squares method to perform surface fitting on the BIM point cloud dataset and the effective LiDAR point cloud data satisfies the following formula: , where, represents the parameter to be optimized, that is, the parameter defining the fitting surface model, represents the number of points participating in the fitting, represents the adaptive weight of the th point, represents the

[0055] fitting surface function. , is the data quality index, including point cloud density, noise level, etc., is the regional importance index, including key components, etc., and are the normalization coefficients. 。

[0056] Furthermore, evaluating the construction quality of highway engineering by combining the registration result and the surface fitting result in step S4 includes the following steps: S41. Calculate the first Euclidean distance between the surface points of each component based on the point cloud data on the fitting surface.

[0057] Specifically, based on the BIM point cloud fitting surface and the LiDAR point cloud fitting surface on the point cloud data, calculate the first Euclidean distance between the pairs of points on the two fitting surfaces of each component point. The calculation of the first Euclidean distance satisfies the following formula: , , , is the point cloud coordinate of the k-th point on the fitting surface , , , is the spatial coordinate of the corresponding point on the fitting surface .

[0058] S42. Generate a deviation data set according to the first Euclidean distance, and use the deviation data set to evaluate the construction quality of highway engineering.

[0059] Adopt the method of local weighted average to generate a deviation data set, map the deviation data set to a red-yellow-green three-color coding map (red: deviation > threshold; yellow: deviation acceptable; green: conforming to the design), generate a deviation heat map, and realize three-dimensional visualization.

[0060] Furthermore, considering the spatial distribution and local characteristics of the point cloud data, perform weighted average on the deviation: , is the neighborhood point set of point , is the original deviation value of point , is the weight of the neighborhood point to point , , represents the Euclidean distance between point and point .

[0061] Furthermore, overlay the deviation heat map and the stage completion data on the BIM model to construct a dynamic visualization platform to display the construction quality in real time, that is:

[0062] is a fusion function, represents the BIM model, represents the deviation heat map, represents the stage completion degree.

[0063] Furthermore, a table containing statistical information such as the maximum deviation, average deviation, and over-limit area of the structure and terrain is generated to visually display the construction quality.

[0064] Furthermore, the number and area of point clouds with deviations exceeding a preset threshold are statistically analyzed to evaluate indicators such as the flatness and perpendicularity of the construction surface, and to determine whether there are out-of-tolerance or defects.

[0065] Furthermore, real-time viewing and interaction of the comprehensive report are realized through the Web end and mobile applications. The deviation data is associated with the construction log, and a warning threshold is set. When the deviation exceeds the threshold, a warning notice is automatically triggered.

[0066] Please refer to Figure 2 , in the embodiment, in order to efficiently execute a method for rapid global perception of highway engineering construction quality provided by the present invention, the present invention also provides a system for rapid global perception of highway engineering construction quality, including: an input device, an output device, a processor, and a memory. The input device, output device, processor, and memory are interconnected. The memory contains program instructions, and the program instructions are used for the steps of the method for rapid global perception of highway engineering construction quality. The system for rapid global perception of highway engineering construction quality of the present invention has a compact structure and stable performance, and can stably execute the method for rapid global perception of highway engineering construction quality of the present invention, further improving the overall applicability and practical application ability of the present invention.

[0067] In an embodiment, the so-called processor may be a Central Processing Unit (CPU), and the processor may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The input device may be used to obtain data information. The output device may be used to output the result obtained from the program instructions included in the computer program stored in the memory provided by the present invention. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A part of the memory may also include a non-volatile random access memory.

[0068] In a possible implementation manner, the memory may include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function, etc.; the data storage area may store the data created during use. In addition, the memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A part of the memory may also include NVRAM. The memory stores an operating system, operation instructions, executable modules, or data structures, or subsets thereof, or extended sets thereof. Among them, the operation instructions may include various operation instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and processing hardware-based tasks.

[0069] The embodiment also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for rapidly perceiving the overall quality of highway engineering construction are implemented.

[0070] The storage medium may include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs.

[0071] In summary, the present invention solves the problem of rapidly quantifying construction quality by fusing and comparing the BIM design model and the actual point cloud data of highway engineering, and then quantitatively analyzing the deviation between the actual construction state and the design model, providing real-time and reliable data support for construction management.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present invention, and they should all be covered within the scope described in the present invention.

Claims

1. A method for rapidly perceiving the overall quality of highway engineering construction, characterized in that, Including the following steps: Construct a BIM design model for the highway project, sample the BIM design model to form a BIM point cloud data set; Obtain the LiDAR point cloud data of the construction area, clean the LiDAR point cloud data to obtain effective LiDAR point cloud data; Perform initial spatial alignment and adaptive fine registration on the BIM point cloud data set and the effective LiDAR point cloud data; Use the least squares method to perform surface fitting on the BIM point cloud data set and the effective LiDAR point cloud data, and evaluate the construction quality of the highway project by combining the registration results and the surface fitting results.

2. The method for rapid perception of the overall construction quality of highway engineering according to claim 1, wherein The sampling of the BIM design model to form a BIM point cloud data set includes the following steps: Convert the BIM design model into a triangular mesh and uniformly sample from the triangular faces; Based on the sampling results, generate BIM point clouds according to the centroids of the triangles.

3. The method for rapidly and comprehensively perceiving the construction quality of highway engineering according to claim 1, characterized in that, The data cleaning of the LiDAR point cloud data to obtain effective LiDAR point cloud data includes the following steps: Perform erosion and dilation operations on the LiDAR point cloud data; Set an elevation difference threshold model, and perform opening operation on the erosion and dilation operation results according to the elevation difference threshold to obtain the effective LiDAR point cloud data.

4. The method for rapid global perception of highway engineering construction quality according to claim 1, wherein The initial spatial alignment of the BIM point cloud data set and the effective LiDAR point cloud data includes the following steps: Segment the BIM point cloud data set and the effective LiDAR point cloud data according to the elevation characteristics and point cloud density differences; Construct an adaptive threshold model, and extract a feature point set from the segmented BIM point cloud data set and effective LiDAR point cloud data through the adaptive threshold model; Calculate the rigid body transformation matrix using the feature point set to complete the initial spatial alignment.

5. The method for rapidly perceiving the overall quality of highway engineering construction according to claim 4, characterized in that, The adaptive threshold model satisfies the following formula: ， Among them, represents the adaptive threshold, represents the mean value of the local curvatures of all points in the point cloud, represents the adjustment coefficient, which can be adjusted according to the actual situation of the point cloud data, represents the standard deviation of the local curvature values of all points in the point cloud.

6. The method for rapid perception of the overall construction quality of highway engineering according to claim 1, wherein The adaptive fine registration of the BIM point cloud data set and the effective LiDAR point cloud data includes the following steps: Construct a fine registration error model and a fine registration objective function; Use the fine registration error model to adjust the weight coefficients in the fine registration objective function, and obtain an accurate rigid body transformation matrix through the adjusted fine registration objective function to complete the adaptive fine registration.

7. The method for rapidly perceiving the overall quality of highway engineering construction according to claim 6, wherein, The fine registration error model satisfies the following formula: ， Among them, represents the registration error, represents the number of point clouds participating in the registration, represents the rotation matrix of i.e., the coordinate system rotation from the BIM point cloud to the LiDAR point cloud, represents the coordinate of the th three-dimensional point in the BIM point cloud, represents the translation vector of i.e., the coordinate system translation from the BIM point cloud to the LiDAR point cloud, represents the coordinate of the th three-dimensional point in the LiDAR point cloud.

8. The method for rapid perception of the overall construction quality of highway engineering according to claim 6, wherein The fine registration objective function satisfies the following formula: ， Among them, represents the value of the fine registration objective function, represents the rotation matrix of which represents the coordinate system rotation from the BIM point cloud to the LiDAR point cloud, the translation vector of which represents the coordinate system translation from the BIM point cloud to the LiDAR point cloud, represents the total number of points in the BIM point cloud, represents the total number of points in the LiDAR point cloud, represents the BIM point and the weight coefficient between the LiDAR points represents the th three-dimensional point coordinate in the BIM point cloud, represents the th three-dimensional point coordinate in the LiDAR point cloud.

9. The method for rapid perception of the overall construction quality of highway engineering according to claim 1, wherein, The evaluation of the construction quality of the highway project by combining the registration results and the surface fitting results includes the following steps: Based on the point cloud data on the fitted surface, calculate the first Euclidean distance between the surface points of each component; Generate a deviation data set according to the first Euclidean distance, and use the deviation data set to evaluate the construction quality of the highway project.

10. A highway engineering construction quality global rapid perception system, characterized in that, The full-region rapid perception system for the construction quality of the highway project includes: an input device, an output device, a processor, and a memory. The input device, the output device, the processor, and the memory are interconnected. The memory includes program instructions, and the program instructions are used to execute the full-region rapid perception method for the construction quality of the highway project according to any one of claims 1-9.

Images