Method for constructing three-dimensional model of underground pipe network

Abstract

The present invention relates to the technical field of three-dimensional construction of pipeline networks, and specifically to a method for constructing a three-dimensional model of an underground pipeline network. The present invention analyzes the distribution of point cloud data of each cubic grid area at different times, and obtains the importance of each cubic grid area based on the complexity of the progressive distribution of the pipeline network in each cubic grid area, the initial amplified pipeline network expression at the real-time moment, and the amount of point cloud data; combines a preset initial neighborhood radius and a preset adjustable neighborhood radius to obtain a corrected neighborhood radius of the point cloud data corresponding to each cubic grid area; clusters the point cloud data in the corresponding cubic grid area to obtain multiple pipeline point cloud clusters, and screens out effective pipeline point cloud clusters; and constructs a three-dimensional model of the underground pipeline network based on the point cloud data of all effective pipeline point cloud clusters. The present invention optimizes the accuracy of point cloud data and improves the accuracy of the three-dimensional model by obtaining an accurate neighborhood radius when clustering point cloud data.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional construction of pipe networks, and in particular to a three-dimensional model construction method for underground pipe networks. Background Art

[0002] With the advancement of information technology, especially the application of big data, building information modeling (BIM), and geographic information systems (GIS), three-dimensional modeling of underground pipeline networks has gradually become an important means of urban planning, pipeline network management, and maintenance. When constructing a three-dimensional model of an underground pipeline network, it is usually necessary to first collect its point cloud data and then denoise it. This is because the collected point cloud data is often interfered by noise, resulting in a decrease in accuracy. Therefore, denoising the point cloud data is essential.

[0003] In the existing technology, the DBSCAN algorithm is used to classify point cloud data. Points with too low density or isolated points can be removed to improve the overall accuracy. However, since the traditional DBSCAN algorithm usually uses a fixed neighborhood radius size during processing, it cannot be adaptively adjusted according to the actual distribution of point cloud positions, which affects the clustering results and denoising effect of the point cloud data, and the accuracy of 3D model construction is poor. Summary of the Invention

[0004] In order to solve the technical problem that a fixed neighborhood radius cannot be adaptively adjusted according to the distribution of actual point cloud positions, resulting in poor accuracy in 3D model construction, the present invention aims to provide a 3D model construction method for underground pipe networks. The technical solutions adopted are as follows:

[0005] The present invention proposes a method for constructing a three-dimensional model of an underground pipe network, the method comprising:

[0006] Obtain point cloud data of the underground pipe network area at every moment;

[0007] The point cloud data is divided into multiple cubic grid areas. Based on the distribution of point cloud data in each cubic grid area at different times, the pipe network progressive distribution complexity of each cubic grid area is obtained, and the primary augmented pipe network representation of each cubic grid area at real time is obtained. Based on the number of point cloud data in each cubic grid area, the pipe network progressive distribution complexity, and the primary augmented pipe network representation, the importance of each cubic grid area is obtained.

[0008] Obtain the preset initial neighborhood radius and preset adjustable neighborhood radius of each cubic grid area, and combine the importance of each cubic grid area to obtain the corrected neighborhood radius of the point cloud data corresponding to each cubic grid area; cluster the point cloud data in the corresponding cubic grid area based on the corrected neighborhood radius of the point cloud data corresponding to different cubic grid areas, obtain multiple pipe network point cloud clusters, and screen out valid pipe network point cloud clusters;

[0009] The three-dimensional model of the underground pipeline network is obtained based on the point cloud data of all valid point cloud clusters of the pipeline network.

[0010] Furthermore, the method for obtaining the pipeline network progressive distribution complexity includes:

[0011] The pipeline network progressive distribution complexity is obtained according to the formula for obtaining the pipeline network progressive distribution complexity of each cubic grid area. The formula for obtaining the pipeline network progressive distribution complexity is:

[0012] Among them, Q i represents the progressive distribution complexity of the pipe network in the i-th cubic grid area; |Δq i,m | represents the difference in the number of point cloud data between the mth moment and the m+1th moment in the i-th cube grid area; q i,m Indicates the number of point cloud data in the i-th cube grid area at the m-th moment; N i,m N represents the point cloud data set of the i-th cube grid area at the m-th moment; i,m+1 Represents the point cloud data set of the i-th cube grid area at the m+1th time; N i,m ∩N i,m+1 N represents the number of intersections of the point cloud data sets of the i-th cube grid area at the m-th moment and the m+1-th moment; i,m ∪N i,m+1 It represents the union number of point cloud data sets of the i-th cube grid area at the m-th moment and the m+1-th moment; M represents the number of all moments; norm() represents the normalization function.

[0013] Furthermore, the method for obtaining the primary amplified pipe network performance includes:

[0014] The primary augmented network representation of each cubic grid area is obtained based on the difference in the number of point cloud data between the real time and the previous adjacent time, as well as the number of point cloud data at the previous adjacent time. The difference in number is positively correlated with the primary augmented network representation, while the number of point cloud data at the previous adjacent time is negatively correlated with the primary augmented network representation.

[0015] Furthermore, the method for obtaining the degree of important attention includes:

[0016] According to the number of point cloud data in each cubic grid area, the density representation of each cubic grid area is obtained;

[0017] The product of the density performance of each cubic grid area and the progressive distribution complexity of the pipe network is obtained as the first level of concern; the difference between the positive integer 1 and the density performance of each cubic grid area is calculated, and the product of the difference result and the initial amplified pipe network performance is obtained as the second level of concern;

[0018] The sum of the first attention level and the second attention level is obtained as the key attention level of each cube grid area.

[0019] Furthermore, the method for obtaining the density expression includes:

[0020] The number of point cloud data in each cube grid area is normalized as the density representation of each cube grid area.

[0021] Furthermore, the method for obtaining the modified neighborhood radius includes:

[0022] Obtaining the product of the importance of each cube grid area and the preset adjustable neighborhood radius as the weighted adjustable neighborhood radius;

[0023] The sum of the weighted adjustable neighborhood radius and the preset initial neighborhood radius is obtained as the modified neighborhood radius of each cube grid area.

[0024] Furthermore, the method for obtaining the pipe network point cloud cluster includes:

[0025] Based on the modified neighborhood radius, the DBSCAN algorithm is performed on the point cloud data in each cubic grid area to obtain multiple pipe network point cloud clusters in each cubic grid area.

[0026] Furthermore, the screening of effective point cloud clusters of the pipe network includes:

[0027] The number of point cloud data in each pipe network point cloud cluster is normalized as the normalized point cloud density value of each pipe network point cloud cluster; if the normalized point cloud density value is greater than the preset density threshold, the corresponding pipe network point cloud cluster is regarded as a valid pipe network point cloud cluster.

[0028] Furthermore, the method for obtaining the three-dimensional model of the underground pipe network includes:

[0029] Import the point cloud data of all effective point cloud clusters of the pipeline network into the BIM software to build a three-dimensional model of the underground pipeline.

[0030] Furthermore, the preset density threshold is 0.1.

[0031] The present invention has the following beneficial effects:

[0032] In order to simplify processing and improve processing efficiency, the present invention divides the point cloud data into multiple cubic grid areas. According to the distribution of point cloud data in each cubic grid area at different times, the spatial distribution change of point cloud data in the grid area is reflected, which helps to understand the complexity of the pipe network structure, obtain the progressive distribution complexity of the pipe network in each cubic grid area, and obtain the primary augmented pipe network expression of each cubic grid area at real time, reflecting an intuitive description of the pipe network structure at real time, which helps to identify the changing trend of the pipe network; according to the number of point cloud data in each cubic grid area, the progressive distribution complexity of the pipe network and the primary augmented pipe network expression, the importance of each cubic grid area is obtained. , more comprehensively evaluate the importance of each grid area; obtain the preset initial neighborhood radius and preset adjustable neighborhood radius of each cubic grid area, and combine the importance of each cubic grid area to obtain the corrected neighborhood radius of the point cloud data corresponding to each cubic grid area, which helps to obtain a corrected neighborhood radius that better meets actual needs; according to the corrected neighborhood radius of the point cloud data corresponding to different cubic grid areas, cluster the point cloud data in the corresponding cubic grid area to obtain multiple pipe network point cloud clusters, screen out the effective pipe network point cloud clusters, help identify the main components of the pipe network structure, and remove noise and redundant data; based on the point cloud data of all effective pipe network point cloud clusters, construct a three-dimensional model of the underground pipe network. The present invention optimizes the accuracy of point cloud data and improves the accuracy of the three-dimensional model by obtaining an accurate neighborhood radius when clustering point cloud data. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 A flowchart of a method for constructing a three-dimensional model of an underground pipe network provided by one embodiment of the present invention.

[0035] Figure 2 A flow chart of a method for obtaining an important attention level provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0036] To further illustrate the technical means and effectiveness of the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail a method for constructing a three-dimensional model of an underground pipe network according to the present invention, including its specific implementation, structure, features, and effectiveness. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0037] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0038] The following describes in detail a method for constructing a three-dimensional model of an underground pipe network provided by the present invention with reference to the accompanying drawings.

[0039] See also Figure 1 , which shows a flow chart of a method for constructing a three-dimensional model of an underground pipe network provided by an embodiment of the present invention. The specific method includes:

[0040] Step S1: Obtain the point cloud data of the underground pipe network area at each moment.

[0041] Underground pipeline networks refer to various pipeline systems buried underground, such as water supply networks, drainage networks, and communication networks, which are used to transport and distribute various resources or provide infrastructure services. By constructing a three-dimensional model, the structure and layout of the underground pipeline network can be displayed in three dimensions, providing a clearer understanding of the actual situation of the pipeline network.

[0042] In an embodiment of the present invention, considering that point cloud data can represent the shape and position of an object, in order to accurately construct a three-dimensional model and remove noise interference, it is necessary to analyze the changes in point cloud data at different times; first, through lidar technology, the point cloud data of the underground pipeline area at each moment is obtained, where the point cloud data contains three-dimensional coordinate information.

[0043] It should be noted that, in the embodiment of the present invention, the time for acquiring point cloud data of the underground pipe network area is set by the implementers according to the actual situation, and is not limited or elaborated here.

[0044] Step S2: Divide the point cloud data into multiple cubic grid areas, and obtain the pipe network progressive distribution complexity of each cubic grid area based on the point cloud data distribution of each cubic grid area at different times, and obtain the initial augmented pipe network expression of each cubic grid area at real time; obtain the importance of each cubic grid area based on the amount of point cloud data, the pipe network progressive distribution complexity, and the initial augmented pipe network expression in each cubic grid area.

[0045] Point cloud data consists of a large number of discrete three-dimensional points. By dividing it into cubic grid regions, it can be converted into more regular data, reducing the range of the point cloud processed each time and improving processing efficiency. The point cloud data is divided into multiple cubic grid regions. It should be noted that in one embodiment of the present invention, the size of the cubic grid region is set to u×u×u to divide the point cloud data. In other embodiments of the present invention, the size of the cubic grid region can be set according to specific circumstances and is not limited or elaborated here.

[0046] For areas where multiple underground pipeline renovations or expansions may be carried out in different time stages, in the same cubic grid area, there may be not only historical point cloud data but also point cloud data at the next moment. The more complex the distribution of the underground pipeline network is, the more complex the distribution is. By analyzing the distribution of point cloud data at different moments, the changes in the pipeline network structure over time are reflected, and the progressive distribution complexity of the pipeline network in each cubic grid area is evaluated. Based on the distribution of point cloud data in each cubic grid area at different moments, the progressive distribution complexity of the pipeline network in each cubic grid area is obtained.

[0047] Preferably, in one embodiment of the present invention, the method for obtaining the progressive distribution complexity of the pipeline network includes:

[0048] The pipeline network progressive distribution complexity is obtained according to the formula for obtaining the pipeline network progressive distribution complexity of each cubic grid area. The formula for obtaining the pipeline network progressive distribution complexity is:

[0049]

[0050] Among them, Q i represents the progressive distribution complexity of the pipe network in the i-th cubic grid area; |Δq i,m | represents the difference in the number of point cloud data between the mth moment and the m+1th moment in the i-th cube grid area; q i,m Indicates the number of point cloud data in the i-th cube grid area at the m-th moment; N i,m N represents the point cloud data set of the i-th cube grid area at the m-th moment; i,m+1 Represents the point cloud data set of the i-th cube grid area at the m+1th time; N i,m ∩N i,m+1 N represents the number of intersections of the point cloud data sets of the i-th cube grid area at the m-th moment and the m+1-th moment; i,m ∪N i,m+1 It represents the union number of the point cloud data sets of the i-th cube grid area at the m-th moment and the m+1-th moment; M represents the number of all moments; norm() represents the normalization function; norm() represents the normalization function.

[0051] In the formula for obtaining the complexity of the progressive distribution of the pipeline network, It represents the ratio between the difference in the number of point cloud data in the ith cubic grid area between the mth moment and the m+1th moment and the number of point cloud data in the ith cubic grid area at the mth moment. The larger the ratio, the greater the difference in the number of point cloud data in the ith cubic grid area between the mth moment and the m+1th moment, that is, the greater the change in point cloud data between adjacent moments, the greater the change in the underground pipe network. It represents the ratio of the number of intersections and unions of the point cloud data sets at the mth moment and the m+1th moment in a cubic grid area. The larger the ratio, the more intersections there are, which means that more point cloud data at the mth moment are retained in the grid area. The underground pipeline network changes on the previous basis. The more changes there are, the more new upgrades there are, and the greater the complexity of the progressive distribution.

[0052] As cities develop and plan, the urban planning of different areas will also change accordingly, causing the amount of point cloud data in the area to change. By analyzing the distribution of point cloud data at different times, reflecting the changes in the pipeline network structure, the initial augmented pipeline network representation of each cubic grid area at the real time is obtained.

[0053] Preferably, in one embodiment of the present invention, the method for obtaining the initial augmented pipe network performance includes:

[0054] The primary augmented network representation of each cubic grid area is obtained based on the difference in the number of point cloud data between the real time and the previous adjacent time, as well as the number of point cloud data at the previous adjacent time. The difference in number is positively correlated with the primary augmented network representation, while the number of point cloud data at the previous adjacent time is negatively correlated with the primary augmented network representation.

[0055] Among them, the larger the quantity difference, the greater the difference in point cloud data between each cube grid area at the real time and the previous adjacent moment. The larger the point cloud data of each cube grid area at the real time, the greater the possibility of expansion, which is a positive correlation; the smaller the number of point cloud data at the previous adjacent moment, the more it indicates that it was close to an undeveloped area before, and the greater the possibility that the cube grid area is the first expansion, which is a negative correlation.

[0056] In one embodiment of the present invention, the formula for the primary amplification type pipe network performance is expressed as:

[0057]

[0058] Among them, W i represents the initial augmented pipe network performance of the i-th cubic grid area; Δq irepresents the difference in the number of point cloud data between the real time and the previous adjacent time in the i-th cube grid area; q ′ i Represents the number of point cloud data in the i-th cube grid area at the previous adjacent moment; norm() represents the normalization function.

[0059] In the formula for the performance of the primary amplified pipe network, It represents the ratio of the difference in the number of point cloud data between the real time and the previous adjacent time in the i-th cube grid area to the number of point cloud data at the previous adjacent time. The larger the ratio, the greater the difference in the number of point cloud data between the real time and the previous adjacent time, and the more point cloud data at the real time, indicating that the underground pipe network distribution at the real time is more complex. The smaller the number of point cloud data at the previous adjacent time, the simpler the underground pipe network distribution at the previous moment, and the more likely it is that it has not been developed. Therefore, the larger the ratio, the greater the representation of the primary augmented pipe network.

[0060] The number of point cloud data reflects the level of detail and complexity of the area. An area with more point cloud data contains more geometric information and details, and requires more refined processing. The progressive distribution complexity of the pipeline network describes the degree of complexity of the pipeline network structure transformation at multiple times. The greater the progressive distribution complexity of the pipeline network, the more attention it requires. The initial augmented pipeline network expression reflects the possibility that the pipeline network has just been formed and requires special attention. Based on the number of point cloud data, the progressive distribution complexity of the pipeline network, and the initial augmented pipeline network expression in each cubic grid area, the importance of each cubic grid area is obtained.

[0061] Preferably, in one embodiment of the present invention, the method for obtaining the important attention level is as follows: Figure 2 , which shows a flow chart of a method for obtaining an important level of attention, including:

[0062] Step S201: Obtaining the density representation of each cubic grid area according to the amount of point cloud data in each cubic grid area.

[0063] The number of point cloud data reflects the density of the point cloud in the cube grid area. The larger the number of point clouds, the greater the density.

[0064] Preferably, in an embodiment of the present invention, the density expression is obtained by normalizing the number of point cloud data of each cubic grid area to obtain the density expression of each cubic grid area.

[0065] It should be noted that the number of point cloud data in each cubic grid area is normalized, that is, the maximum number of point cloud data in all cubic grid areas is obtained, and the ratio between the number of point cloud data in each cubic grid area and the maximum number of point cloud data is obtained as the density representation of each cubic grid area.

[0066] Step S202: Obtain the product of the density expression of each cubic grid area and the progressive distribution complexity of the pipe network as the first level of concern; calculate the difference between the positive integer 1 and the density expression of each cubic grid area, and obtain the product of the difference result and the initial augmented pipe network expression as the second level of concern.

[0067] The greater the density, the more likely it is that the current grid area has complex pipe network distributions at different times. Therefore, more emphasis should be placed on the complex progressive distribution of the pipe network in the grid area. Therefore, the density expression is used as the weight of the complexity of the progressive distribution of the pipe network. The smaller the density, the more likely it is that the grid area has a initially expanded pipe network distribution. More attention should be paid to the initially expanded pipe network. Therefore, the difference between the positive integer 1 and the density expression of each cubic grid area is calculated as the weight of the initially expanded pipe network expression.

[0068] Step S203: obtaining the sum of the first attention level and the second attention level as the key attention level of each cubic grid area.

[0069] In one embodiment of the present invention, the formula for the degree of important attention is expressed as:

[0070] R i =E i ×Q i +(1-E i )×W i ;

[0071] Among them, R i Indicates the importance of the i-th cube grid area; E i represents the density expression of the i-th cube grid area; Q i represents the progressive distribution complexity of the pipe network in the i-th cubic grid area; W i Represents the primary augmented pipe network representation of the i-th cube grid area.

[0072] In the formula of the degree of importance of attention, the greater the density expression of the i-th cube grid area, the more it indicates that the grid area may have complex pipeline network distribution at different times; the higher the attention to the complexity of the progressive distribution of the pipeline network, the smaller the density expression, the more attention should be paid to the possible initial expansion of the pipeline network distribution in the grid area, and the higher the Gu'an Pig degree of the initial expansion type pipeline network expression; the greater the complexity of the progressive distribution of the pipeline network, the greater the initial expansion type pipeline network expression, and the greater the degree of importance of attention.

[0073] Step S3: Obtain the preset initial neighborhood radius and the preset adjustable neighborhood radius of each cubic grid area, and obtain the corrected neighborhood radius of the point cloud data corresponding to each cubic grid area in combination with the importance of each cubic grid area; cluster the point cloud data in the corresponding cubic grid area according to the corrected neighborhood radius of the point cloud data corresponding to different cubic grid areas, obtain multiple pipe network point cloud clusters, and screen out valid pipe network point cloud clusters.

[0074] Because the distribution density and shape of point clouds may vary across different cubic grid regions, the neighborhood radius needs to be modified to accommodate different local spatial characteristics. The initial neighborhood radius provides a basic range for local searches of point cloud data, ensuring that a sufficient number of neighboring points can be found during neighborhood searches to construct spatial topological relationships between point cloud data. The adjustable neighborhood radius allows for adjustment of the initial neighborhood radius based on specific circumstances. The importance of each cubic grid region reflects the relative importance of that region in the overall point cloud data. Weighting the adjustable neighborhood radius based on the importance of each cubic grid region ensures more accurate retrieval of relevant point cloud data during the search process, helping to reduce the impact of noise and redundant data and improve the accuracy of search results. The preset initial neighborhood radius and preset adjustable neighborhood radius for each cubic grid region are obtained, and combined with the importance of each cubic grid region, the modified neighborhood radius for the point cloud data corresponding to each cubic grid region is obtained.

[0075] Preferably, in one embodiment of the present invention, the method for obtaining the modified neighborhood radius includes:

[0076] Obtaining the product of the importance of each cube grid area and the preset adjustable neighborhood radius as the weighted adjustable neighborhood radius;

[0077] The sum of the weighted adjustable neighborhood radius and the preset initial neighborhood radius is obtained as the modified neighborhood radius of each cube grid area.

[0078] In one embodiment of the present invention, the formula for correcting the neighborhood radius is expressed as:

[0079] T i =k1+R i×k2;

[0080] Among them, T i represents the modified neighborhood radius of the i-th cube grid area; k1 represents the preset initial neighborhood radius; R i Indicates the importance of the i-th cube grid area; k2 represents the preset adjustable neighborhood radius.

[0081] In the formula for correcting the neighborhood radius, R i ×k2 represents the product of the importance of the i-th cubic grid area and the preset adjustable neighborhood radius. That is, the preset adjustable neighborhood radius is weighted by the importance of the i-th cubic grid area. As the weighted adjustable neighborhood radius, the greater the importance, the more details need to be retained. The larger the weighted adjustable neighborhood radius, the larger the corrected neighborhood radius.

[0082] It should be noted that in the embodiments of the present invention, the implementer can obtain the neighborhood radius adjustment range based on the empirical threshold to obtain the initial neighborhood radius and the adjustable neighborhood radius. For example, if the neighborhood radius adjustment range is 3-12, the initial neighborhood radius is preset to 3 and the adjustable neighborhood radius is preset to 9. The adjustable neighborhood radius is weighted according to the degree of importance to obtain a weighted adjustable neighborhood radius. The specific means are well known to those skilled in the art and will not be elaborated here.

[0083] In point cloud processing, clustering algorithms can divide point cloud data with similar spatial distribution or attributes into the same cluster, which helps to identify different pipe network components and reduce the impact of noise points or abnormal points on clustering results; by correcting the neighborhood radius, the spatial relationship between point cloud data can be more accurately described, thereby improving the accuracy of clustering analysis; according to the corrected neighborhood radius of the point cloud data corresponding to different cubic grid areas, the point cloud data in the corresponding cubic grid area are clustered to obtain multiple pipe network point cloud clusters, and the effective pipe network point cloud clusters are screened out.

[0084] Preferably, in one embodiment of the present invention, the method for obtaining a pipe network point cloud cluster includes:

[0085] Based on the modified neighborhood radius, the DBSCAN algorithm is performed on the point cloud data in each cubic grid area to obtain multiple pipe network point cloud clusters in each cubic grid area.

[0086] The DBSCAN clustering algorithm can effectively identify point cloud data within a neighborhood radius and divide it into clusters while eliminating the influence of noise points. It should be noted that in the DBSCAN clustering algorithm, the size of the neighborhood radius parameter determines the density of the cluster, and the minimum number of samples is set to 20, that is, only when the cluster corresponding to the point cloud data contains at least 20 point cloud data can the clustering condition be met; in other embodiments of the present invention, the size of the minimum number of samples can be set according to the specific situation, and will not be limited or elaborated here. The specific DBSCAN clustering is a technical means well known to those skilled in the art and will not be elaborated here.

[0087] Preferably, in one embodiment of the present invention, screening out effective pipeline point cloud clusters includes:

[0088] The number of point cloud data in each pipe network point cloud cluster is normalized as the normalized point cloud density value of each pipe network point cloud cluster; if the normalized point cloud density value is greater than the preset density threshold, the corresponding pipe network point cloud cluster is regarded as a valid pipe network point cloud cluster.

[0089] It should be noted that, in one embodiment of the present invention, the preset density threshold is 0.1; in other embodiments of the present invention, the size of the preset density threshold may be set according to specific circumstances, which is not limited or elaborated herein.

[0090] Step S4: Construct a three-dimensional model of the underground pipeline network based on the point cloud data of all valid point cloud clusters of the pipeline network.

[0091] After filtering out valid point cloud clusters of the pipeline network, the influence of noise points with too low density on corresponding clusters is avoided, which helps to build a more accurate 3D model. Based on the point cloud data of all valid point cloud clusters of the pipeline network, a 3D model of the underground pipeline network is constructed.

[0092] It should be noted that, in one embodiment of the present invention, after obtaining the point cloud data of all valid point cloud clusters of the pipeline network, a three-dimensional model of the underground network is constructed, including: importing the point cloud data of all valid point cloud clusters of the pipeline network into BIM software to construct a three-dimensional model of the underground network; the specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0093] In summary, the present invention divides point cloud data into multiple cubic grid regions, analyzes the distribution of point cloud data in each cubic grid region at different times, and determines the importance of each cubic grid region based on the progressive distribution complexity of the pipeline network, the real-time representation of the initial augmented pipeline network, and the amount of point cloud data in each cubic grid region. Combining a preset initial neighborhood radius with a preset adjustable neighborhood radius, the modified neighborhood radius of the point cloud data corresponding to each cubic grid region is obtained. The point cloud data within the corresponding cubic grid region is clustered to obtain multiple pipeline network point cloud clusters, which are then screened for valid pipeline network point cloud clusters. A three-dimensional model of the underground pipeline network is constructed based on the point cloud data from all valid pipeline network point cloud clusters. By obtaining an accurate neighborhood radius when clustering point cloud data, the present invention optimizes the precision of the point cloud data and improves the accuracy of the three-dimensional model.

[0094] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0095] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for constructing a three-dimensional model of an underground pipe network, characterized in that: The method comprises: Obtain point cloud data of the underground pipe network area at every moment; The point cloud data is divided into multiple cubic grid areas. Based on the distribution of point cloud data in each cubic grid area at different times, the pipe network progressive distribution complexity of each cubic grid area is obtained, and the primary augmented pipe network representation of each cubic grid area at real time is obtained. Based on the number of point cloud data in each cubic grid area, the pipe network progressive distribution complexity, and the primary augmented pipe network representation, the importance of each cubic grid area is obtained. Obtain the preset initial neighborhood radius and preset adjustable neighborhood radius of each cubic grid area, and combine the importance of each cubic grid area to obtain the corrected neighborhood radius of the point cloud data corresponding to each cubic grid area; cluster the point cloud data in the corresponding cubic grid area based on the corrected neighborhood radius of the point cloud data corresponding to different cubic grid areas, obtain multiple pipe network point cloud clusters, and screen out valid pipe network point cloud clusters; Obtain a three-dimensional model of the underground pipeline network based on the point cloud data of all valid point cloud clusters of the pipeline network; The method for obtaining the pipeline network progressive distribution complexity includes: The pipeline network progressive distribution complexity is obtained according to the formula for obtaining the pipeline network progressive distribution complexity of each cubic grid area. The formula for obtaining the pipeline network progressive distribution complexity is: Among them, Q i represents the progressive distribution complexity of the pipe network in the i-th cubic grid area; |Δq i,m | represents the difference in the number of point cloud data between the mth moment and the m+1th moment in the i-th cube grid area; q i,m Indicates the number of point cloud data in the i-th cube grid area at the m-th moment; N i,m Represents the point cloud data set of the i-th cube grid area at the m-th moment; N i,m+1 Represents the point cloud data set of the i-th cube grid area at the m+1th moment; N i,m ∩N i,m+1 N represents the number of intersections of the point cloud data sets of the i-th cube grid area at the m-th time and the m+1-th time; i,m ∪N i,m+1 It represents the union number of point cloud data sets of the i-th cube grid area at the m-th moment and the m+1-th moment; M represents the number of all moments; norm() represents the normalization function.

2. A method for constructing a three-dimensional model of an underground pipe network according to claim 1, characterized in that: The method for obtaining the primary amplified pipe network performance includes: The primary augmented network representation of each cubic grid area is obtained based on the difference in the number of point cloud data between the real time and the previous adjacent time, as well as the number of point cloud data at the previous adjacent time. The difference in number is positively correlated with the primary augmented network representation, while the number of point cloud data at the previous adjacent time is negatively correlated with the primary augmented network representation.

3. A method for constructing a three-dimensional model of an underground pipe network according to claim 1, characterized in that: The method for obtaining the important attention level includes: According to the number of point cloud data in each cubic grid area, the density representation of each cubic grid area is obtained; The product of the density performance of each cubic grid area and the progressive distribution complexity of the pipe network is obtained as the first level of concern; the difference between the positive integer 1 and the density performance of each cubic grid area is calculated, and the product of the difference result and the initial amplified pipe network performance is obtained as the second level of concern; The sum of the first attention level and the second attention level is obtained as the key attention level of each cube grid area.

4. A method for constructing a three-dimensional model of an underground pipe network according to claim 3, characterized in that: The method for obtaining the density expression comprises: The number of point cloud data in each cube grid area is normalized as the density representation of each cube grid area.

5. The method for constructing a three-dimensional model of an underground pipe network according to claim 1, wherein: The method for obtaining the modified neighborhood radius includes: Obtaining the product of the importance of each cube grid area and the preset adjustable neighborhood radius as the weighted adjustable neighborhood radius; The sum of the weighted adjustable neighborhood radius and the preset initial neighborhood radius is obtained as the modified neighborhood radius of each cube grid area.

6. A method for constructing a three-dimensional model of an underground pipe network according to claim 1, characterized in that: The method for obtaining the pipe network point cloud cluster includes: Based on the modified neighborhood radius, the DBSCAN algorithm is performed on the point cloud data in each cubic grid area to obtain multiple pipe network point cloud clusters in each cubic grid area.

7. A method for constructing a three-dimensional model of an underground pipe network according to claim 1, characterized in that: The filtering out of effective point cloud clusters of the pipe network includes: The number of point cloud data in each pipe network point cloud cluster is normalized as the normalized point cloud density value of each pipe network point cloud cluster; if the normalized point cloud density value is greater than the preset density threshold, the corresponding pipe network point cloud cluster is regarded as a valid pipe network point cloud cluster.

8. The method for constructing a three-dimensional model of an underground pipe network according to claim 1, wherein: The method for obtaining the three-dimensional model of the underground pipe network includes: Import the point cloud data of all effective point cloud clusters of the pipeline network into the BIM software to build a three-dimensional model of the underground pipeline.

9. A method for constructing a three-dimensional model of an underground pipe network according to claim 7, characterized in that: The preset density threshold is 0.1.

Images