Underground pipe network three-dimensional model generation method and system based on data fusion

Abstract

The invention discloses an underground pipe network three-dimensional model generation method and system based on data fusion, and relates to the technical field of underground pipe network three-dimensional models, and the method comprises the following steps: multi-source data collection, data preprocessing, dynamic weight data fusion, and model construction and visualization. According to the method, through multi-source data collection, the problem that a traditional method is single in data source is solved, and the actual situation of the underground pipe network can be comprehensively reflected. Through the processes of data preprocessing and dynamic weight data fusion, the data accuracy is effectively improved, and then the generation precision of the underground pipe network three-dimensional model is improved. The dynamic weight data fusion considers the timeliness of the data, so that the model can better adapt to the dynamic change condition of the pipe network, and the reliability of the model is enhanced. Through model construction and visualization, the underground pipe network distribution and structure are visually presented, and powerful support is provided for urban planning work and pipe network maintenance and management.

Description

Technical Field

[0001] The present invention relates to the technical field of underground pipe network three-dimensional models, and in particular to a method and system for generating a three-dimensional model of an underground pipe network based on data fusion. Background Art

[0002] With the rapid advancement of urban construction, underground pipe network systems are becoming more and more complex and diverse. Accurately constructing a three-dimensional model of the underground pipe network plays a vital role in urban planning, maintenance and management of the pipe network. The existing underground pipe network modeling methods face many problems, including a relatively single data source, unsatisfactory model accuracy, and inability to adapt well to the dynamic changes of the pipe network. In view of this, a new method is urgently needed to effectively improve the generation accuracy of the underground pipe network three-dimensional model and enhance its reliability. Summary of the invention

[0003] In view of the problems existing in the prior art, the present invention provides a method and system for generating a three-dimensional model of an underground pipe network based on data fusion. The purpose of the present invention is to provide a method and system for generating a three-dimensional model of an underground pipe network based on data fusion, so as to solve the problem raised in the above background technology of how to dig two different pits at one time when interplanting and keep the relative positions of the two plants stable.

[0004] The present invention is implemented in this way: a method for generating a three-dimensional model of an underground pipe network based on data fusion comprises the following steps:

[0005] Multi-source data collection: Use the geographic information system (GIS) to obtain the geographical location data of the underground pipeline network, use the pipeline inspection robot to collect the internal structure data of the pipeline, and use the ground penetrating radar (GPR) to measure the depth information of the underground pipeline.

[0006] Data preprocessing: coordinate transformation of GIS data, converting longitude and latitude coordinates into plane rectangular coordinates suitable for modeling; using mean filtering method to denoise the depth data obtained by GPR to improve data accuracy.

[0007] Dynamic weight data fusion: Dynamically calculate the weight of GIS and GPR data based on the error estimation and timeliness of data collection. Combine the pre-processed GIS coordinate data with GPR depth data to generate the three-dimensional coordinates of the pipeline network nodes, and build a three-dimensional geometric model of the pipeline based on the internal structure data of the pipeline.

[0008] Model construction and visualization: Utilize the integrated pipe network data and the constructed geometric model, build a 3D model of the underground pipe network with the help of 3D modeling software or professional GIS 3D modeling module, and perform rendering and visualization operations to intuitively present the distribution and structure of the underground pipe network.

[0009] Preferably, in the present invention, the GIS data coordinate transformation adopts a specific projection method to convert longitude and latitude coordinates into plane rectangular coordinates. The specific transformation formula is determined according to the selected projection system. For example, in the Gauss-Krüger projection, the target coordinates are calculated through parameters such as the arc length of the central meridian and the radius of curvature of the prime vertical circle.

[0010] Preferably, in the present invention, the depth data obtained by GPR is subjected to mean filtering. For the depth data sequence z gpr (i), the filtered value is calculated by the following formula: where n is the size of the filtering window with an odd number, used to remove data noise.

[0011] Preferably, in the present invention, when constructing the three-dimensional geometric model of the pipeline, for a cylindrical pipeline, based on the pipeline central axis, according to the pipe diameter D, the coordinates of the pipeline surface points are determined using parametric equations to construct the three-dimensional shape of the pipeline. The pipeline surface points are related to the parametric equation of the central axis and are related to the pipe diameter and the angle θ.

[0012] Preferably, in the present invention, in the data fusion step, weights are calculated through data error estimation and acquisition time. Data with small errors and recent acquisition times have higher weights. Among them, when calculating the error weight, based on the GIS data error estimation σ gis and the GPR data error estimation σ gpr , the respective error weights are calculated according to a specific formula.

[0013] Preferably, the formula for calculating the error weight is:

[0014]

[0015] Preferably, when calculating the timeliness weight, according to the time t gis elapsed since the GIS data was collected and the time t gpr elapsed since the GPR data was collected, the timeliness weight is determined using an exponential decay function, where the decay coefficient λ can be adjusted according to the actual situation.

[0016] Preferably, the formula for calculating the timeliness weight is:

[0017] w t-gis =e -λt g i s ;

[0018] w t-gpr =e -λt gpr .

[0019] Preferably, in the present invention, a comprehensive weight is obtained by integrating the error weight and the timeliness weight, and then the attribute value v obtained from the GIS data is weighted by the comprehensive weight gis and the attribute value v obtained from the GPR data gpr are fused to obtain the fused value v fused .

[0020] A three-dimensional model generation system for underground pipe networks based on data fusion includes the following modules:

[0021] Multi-source data acquisition module: Using a geographic information system (GIS) to obtain the geographical location data of underground pipe networks, collecting the internal structure data of pipes through a pipeline inspection robot, and measuring the depth information of underground pipelines using ground penetrating radar (GPR);

[0022] Data preprocessing module: Performing coordinate transformation on the GIS data to convert longitude and latitude coordinates into plane rectangular coordinates suitable for modeling; Using the mean filtering method to denoise the depth data obtained by GPR to improve data accuracy;

[0023] Dynamic weight data fusion module: Dynamically calculating the weights of the two based on the error estimation of GIS and GPR data and the timeliness of data acquisition. Combining the preprocessed GIS coordinate data with the GPR depth data to generate the three-dimensional coordinates of pipe network nodes, and constructing a three-dimensional geometric model of the pipeline based on the internal structure data of the pipeline;

[0024] Model construction and visualization module: Using the fused pipe network data and the constructed geometric model, building a three-dimensional model of the underground pipe network with the help of three-dimensional modeling software or a professional GIS three-dimensional modeling module, and performing rendering and visualization operations to intuitively present the distribution and structure of the underground pipe network.

[0025] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0026] Improve data diversity: By multi-source data acquisition, the problem of single data source in traditional methods is overcome, and the actual situation of underground pipe networks can be comprehensively reflected.

[0027] Improve model accuracy: In the data preprocessing and dynamic weight data fusion processes, the data accuracy is effectively improved, and thus the generation accuracy of the three-dimensional model of the underground pipe network is improved.

[0028] Enhance adaptability: The dynamic weight data fusion takes into account the timeliness of data, enabling the model to better adapt to the dynamic changes of the pipe network and enhancing the reliability of the model.

[0029] Facilitate management and planning: Through model construction and visualization, the distribution and structure of the underground pipe network are intuitively presented, providing strong support for urban planning work, pipe network maintenance, and management. Brief Description of the Drawings

[0030] Figure 1 FIG. 1 is a schematic flow chart of the steps of the method for generating a three-dimensional model of an underground pipe network based on data fusion provided by an embodiment of the present invention;

[0031] Figure 2 FIG. 2 is a schematic structural diagram of a system for generating a three-dimensional model of an underground pipe network based on data fusion provided by an embodiment of the present invention. Detailed Embodiments

[0032] In order to further understand the content, features and effects of the present invention, the following embodiments are exemplified and described in detail in conjunction with the accompanying drawings.

[0033] The structure of the present invention will be described in detail below with reference to the accompanying drawings.

[0034] As Figure 1-2 shown, the method for generating a three-dimensional model of an underground pipe network based on data fusion provided by an embodiment of the present invention includes the following steps:

[0035] Multi-source data collection: Use a Geographic Information System (GIS) to obtain the geographical location data of the underground pipe network, collect the internal structure data of the pipeline through a pipeline inspection robot, and measure the depth information of the underground pipeline using a Ground Penetrating Radar (GPR). The Geographic Information System (GIS) can provide the location distribution of the underground pipe network on the map, such as the specific coordinate position of the pipe network on a certain urban street. The pipeline inspection robot can enter the pipeline interior to collect structural data such as the diameter and wall thickness of the pipeline. The Ground Penetrating Radar measures the depth of the underground pipeline from the ground by emitting and receiving electromagnetic waves.

[0036] For example: Conduct underground pipe network data collection in the central area of a certain city. Use GIS to obtain the geographical location data of a sewage pipe network, with the starting point longitude and latitude coordinates being (30.67°N, 104.06°E) and the ending point longitude and latitude coordinates being (30.675°N, 104.065°E). Use a pipeline inspection robot to detect a circular water supply pipeline with a diameter of 0.5 meters and collect a wall thickness of 0.03 meters. Use a Ground Penetrating Radar to measure the depth of a power cable and obtain a depth data of 1.2 meters.

[0037] Data preprocessing: Perform coordinate conversion on the GIS data to convert the longitude and latitude coordinates into plane rectangular coordinates suitable for modeling; use the mean filtering method to denoise the depth data obtained by GPR to improve data accuracy. Since modeling software usually uses plane rectangular coordinates, it is necessary to convert the longitude and latitude coordinates obtained by GIS. The mean filtering method processes the depth data sequence to remove noise interference and make the data more accurate.

[0038] For example: For the latitude and longitude coordinates of the sewage pipe network obtained by the above GIS (30.67°N, 104.06°E), assuming that the Gauss-Krüger projection is used for coordinate transformation, the longitude of the central meridian of this area is 105°. Through the Gauss-Krüger projection formula, the calculated plane rectangular coordinates are (X = 3,400,000 meters, Y = 500,000 meters) (this data is a hypothetical calculation result for demonstration purposes only). For the depth data sequence [1.2, 1.15, 1.22, 1.18, 1.25] meters obtained by the ground penetrating radar, the mean filtering method is adopted, and the size of the filtering window n = 3 (odd). According to the formula When i = 3, After filtering, the data sequence is [1.175, 1.183, 1.21, 1.21, 1.225] meters, and the data accuracy has been improved.

[0039] Dynamic weight data fusion: Dynamically calculate the weights of the two according to the error estimation of GIS and GPR data and the timeliness of data acquisition. Combine the preprocessed GIS coordinate data and GPR depth data to generate the three-dimensional coordinates of the pipe network nodes, and construct the three-dimensional geometric model of the pipeline based on the internal structure data of the pipeline.

[0040] Note: Data error estimation and acquisition timeliness affect the reliability of data. By calculating weights, data from different sources can be reasonably fused, and then the three-dimensional coordinates of the pipe network nodes can be generated and the three-dimensional geometric model of the pipeline can be constructed.

[0041] For example: Assume the error estimation of GIS data meters, and the error estimation of GPR data meters. The time elapsed since the GIS data was collected days, and the time elapsed since the GPR data was collected days, and the attenuation coefficient .

[0042] First, calculate the error weight, ; .

[0043] Then, calculate the timeliness weight, ; ; The comprehensive weight , . Assume the coordinate attribute value of a certain pipe network node obtained from GIS data meters, meters, meters The depth attribute value of this node obtained from the data meters , the three-dimensional coordinates of the node after fusion The value is m. For the construction of the three-dimensional geometric model of the pipeline, if it is a cylindrical water supply pipeline, the pipe diameter m, the parametric equation of the central axis is set as , the points on the pipe surface are related to the parametric equation of the central axis. Using the parametric equation such as to determine the coordinates of the points on the pipe surface and construct the three-dimensional shape of the pipeline.

[0044] Model construction and visualization: Using the fused pipe network data and the constructed geometric model, build a three-dimensional model of the underground pipe network with the help of three-dimensional modeling software or a professional GIS three-dimensional modeling module, and perform rendering and visualization operations to intuitively present the distribution and structure of the underground pipe network. Input the fused data and the constructed geometric model obtained in the previous steps into the three-dimensional modeling software or the professional GIS three-dimensional modeling module, and through operations such as rendering, display the distribution and specific structure of the underground pipe network in an intuitive way.

[0045] For example: Using the three-dimensional modeling module of ArcGIS, import the three-dimensional coordinate data of the above-mentioned fused pipe network nodes and the data of the constructed three-dimensional geometric model of the pipeline. Set parameters such as materials and lighting for rendering. In the three-dimensional view, the distribution of underground pipe networks such as sewage pipe networks, water supply pipelines, and power cables can be intuitively seen. For example, the sewage pipe network is at a depth of 1.5 m underground and is laid along a certain street, and the water supply pipeline is laid parallel 0.3 m above the sewage pipe network, clearly presenting the distribution and structure of the underground pipe network. GIS data coordinate conversion: The GIS data coordinate conversion adopts a specific projection method to convert longitude and latitude coordinates into plane rectangular coordinates. The specific conversion formula is determined according to the selected projection system. For example, for the Gauss-Krüger projection, the target coordinates are calculated through parameters such as the arc length of the central meridian and the radius of curvature of the prime vertical. Different projection methods have different formulas. Taking the Gauss-Krüger projection as an example, parameters such as the arc length of the central meridian and the radius of curvature of the prime vertical are required to calculate the plane rectangular coordinates. For example: As mentioned above, the longitude and latitude coordinates of the starting point of the sewage pipe network (north latitude , east longitude ). Assuming that the Gauss-Krüger projection is adopted in this area and the longitude of the central meridian is . According to the Gauss-Krüger projection formula, first calculate some intermediate parameters, such as the radius of curvature of the prime vertical (where is the semi-major axis of the earth, is the first eccentricity, is the latitude), assuming m, , the calculated result is m. The arc length of the central meridian is also calculated through the corresponding formula (the specific calculation process is omitted here). After a series of calculations, the plane rectangular coordinates of this point are ( m, m) (this data is a hypothetical calculation result for illustration only).

[0046] Perform mean filtering on the depth data obtained by GPR: For the mean filtering of the depth data obtained by GPR, for the depth data sequence z gpr (i), the filtered value is calculated through the following formula: where n is the size of the filtering window with an odd number, used to remove data noise. Processing the depth data sequence according to this formula can remove noise and make the data smoother and more accurate.

[0047] For example: For the depth data sequence [1.2, 1.15, 1.22, 1.18, 1.25] m obtained by ground penetrating radar, take the filtering window size n = 3 (odd). When i = 1, When i = 2, And so on. The filtered data sequence is [1.19, 1.183, 1.21, 1.21, 1.225] m. Compared with the original data, the noise has been effectively removed.

[0048] Construct a three-dimensional geometric model of the pipeline: When constructing a three-dimensional geometric model of the pipeline, for a cylindrical pipeline, based on the pipeline central axis, according to the pipe diameter D, use the parametric equation to determine the coordinates of the pipeline surface points and construct the three-dimensional shape of the pipeline. The pipeline surface points are related to the parametric equation of the central axis and are related to the pipe diameter and the angle θ. Through the parametric equation of the central axis and information such as the pipe diameter, the coordinates of each point on the pipeline surface can be determined using the parametric equation, thereby constructing a three-dimensional shape.

[0049] For example: For a cylindrical water supply pipeline with a pipe diameter D = 0.5 m, assume the parametric equation of the central axis is (s is the distance parameter along the central axis). Using the parametric equation Z = Z(s), when s = 0, θ = 0, the coordinates of the pipeline surface points are When s = 0, θ = 90°, the coordinates of the pipeline surface points are The coordinates of the entire pipeline surface can be determined through different s and θ values, and the three-dimensional shape of the pipeline can be constructed.

[0050] Calculating weights in the data fusion step: In the data fusion step, weights are calculated based on data error estimation and acquisition time. Data with smaller errors and closer acquisition times have higher weights. When calculating the error weights, the error estimations of GIS data, denoted as σ gis and GPR data, denoted as σ gpr , are used to calculate their respective error weights according to a specific formula. Data with smaller errors are more reliable, and data with closer acquisition times have better timeliness. The formula for calculating weights reflects this difference.

[0051] For example: Assume that the error estimation of GIS data, σ gis = 0.05 meters, and the error estimation of GPR data, σ gpr = 0.1 meters. According to the formula , it can be seen that the error weight of GIS data is higher because its error estimation is smaller.

[0052] 9. Formula for calculating error weights: The formula for calculating error weights is:

[0053]

[0054] This formula calculates the respective error weights based on the error estimation values of the data, reflecting the principle that the smaller the error, the larger the weight.

[0055] For example: As in the previous example, when σ gis = 0.05 meters and σ gpr = 0.1 meters, intuitively demonstrates the calculation process and results of the formula.

[0056] Calculating timeliness weights: When calculating timeliness weights, based on the time t gis elapsed since the acquisition of GIS data and the time t gpr elapsed since the acquisition of GPR data, an exponential decay function is used to determine the timeliness weights, where the decay coefficient λ can be adjusted according to the actual situation. The exponential decay function can reflect the characteristic that the timeliness of data decreases over time, and the decay coefficient can be adjusted according to different application scenarios.

[0057] For example: Assume that the time t gis elapsed since the acquisition of GIS data is 2 days, the time t gpr elapsed since the acquisition of GPR data is 5 days, and the decay coefficient λ = 0.1. According to the formula w t-gis = e -0.1×2 ≈ 0.819; w t-gpr = e -0.1×5≈0.607, indicating that due to the more recent data collection time of GIS data, the timeliness weight is higher.

[0058] 11. Formula for calculating the timeliness weight: The formula for calculating the timeliness weight is as follows: Through these two formulas, the timeliness weights of their respective data are calculated based on the data collection time and the attenuation coefficient.

[0059] For example: As in the above example, when t gis = 2 days, t gpr = 5 days, and λ = 0.1, w t-gis = e -0.1×2 ≈0.819, w t-gpr = e -0.1×5 ≈0.607, clearly showing the application of the formula and the calculation results.

[0060] Combined weight and data fusion: The combined error weight and timeliness weight are combined to obtain the combined weight, and then the attribute value v gis obtained from GIS data and the attribute value v gpr obtained from GPR data are fused to obtain.

[0061] Refer to Figure 2 , A three-dimensional model generation system for underground pipe networks based on data fusion, including the following modules:

[0062] Multi-source data acquisition module: Using the Geographic Information System (GIS) to obtain the geographical location data of the underground pipe network, collecting the internal structure data of the pipeline through a pipeline inspection robot, and measuring the depth information of the underground pipeline using Ground Penetrating Radar (GPR);

[0063] Data preprocessing module: Performing coordinate transformation on the GIS data to convert the longitude and latitude coordinates into plane rectangular coordinates suitable for modeling; Using the mean filtering method to denoise the depth data obtained by GPR to improve data accuracy;

[0064] Dynamic weight data fusion module: Dynamically calculating the weights of both based on the error estimation of GIS and GPR data and the timeliness of data collection; Combining the preprocessed GIS coordinate data with the GPR depth data to generate the three-dimensional coordinates of the pipe network nodes, and constructing the three-dimensional geometric model of the pipeline based on the internal structure data of the pipeline;

[0065] Model construction and visualization module: Using the fused pipe network data and the constructed geometric model, and relying on three-dimensional modeling software or professional GIS three-dimensional modeling modules to build a three-dimensional model of the underground pipe network, and performing rendering and visualization operations to intuitively present the distribution and structure of the underground pipe network.

[0066] The working principle of the present invention:

[0067] Multi-source data collection: Use geographic information system (GIS) to obtain geographical location data of underground pipelines, use pipeline inspection robots to collect internal structure data of pipelines, and use ground penetrating radar (GPR) to measure the depth information of underground pipelines. Multiple data sources complement each other and provide a rich data foundation for comprehensive modeling.

[0068] Data preprocessing: Coordinate conversion of GIS data is performed to convert longitude and latitude coordinates into plane rectangular coordinates suitable for modeling, ensuring that data is processed under a unified coordinate system; mean filtering method is used to denoise the depth data obtained by GPR, remove interference signals, and improve data accuracy.

[0069] Dynamic weight data fusion: Dynamically calculate the weight of GIS and GPR data based on the error estimation and timeliness of data collection. Comprehensively consider the reliability and real-time performance of different data sources, combine the pre-processed GIS coordinate data with GPR depth data, generate the three-dimensional coordinates of the pipeline network nodes, and build the three-dimensional geometric model of the pipeline based on the internal structure data of the pipeline.

[0070] Model construction and visualization: Utilize the integrated pipe network data and the constructed geometric model, build a 3D model of the underground pipe network with the help of 3D modeling software or professional GIS 3D modeling module, and perform rendering and visualization operations to intuitively present the distribution and structure of the underground pipe network.

[0071] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device 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, method, article or device.

[0072] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for generating a three-dimensional model of an underground pipe network based on data fusion, characterized in that: The following steps are involved: Multi-source data collection: Use geographic information system GIS to obtain geographical location data of underground pipelines, use pipeline inspection robots to collect internal structure data of pipelines, and use ground penetrating radar GPR to measure the depth information of underground pipelines; Data preprocessing: coordinate conversion of GIS data, converting longitude and latitude coordinates into plane rectangular coordinates suitable for modeling; The depth data acquired by GPR is denoised using the mean filtering method; Dynamic weighted data fusion: Based on the error estimation of GIS and GPR data and the timeliness of data collection, the weights of the two are dynamically calculated, and the 3D coordinates of the pipeline network nodes are generated by combining the preprocessed GIS coordinate data and GPR depth data. The 3D geometric model of the pipeline is constructed based on the internal structure data of the pipeline. Model construction and visualization: Utilize the integrated pipe network data and the constructed geometric model, build a 3D model of the underground pipe network with the help of 3D modeling software or professional GIS 3D modeling module, and perform rendering and visualization operations to intuitively present the distribution and structure of the underground pipe network.

2. The method for generating a three-dimensional model of an underground pipe network based on data fusion according to claim 1, characterized in that: The GIS data coordinate conversion adopts a projection method to convert the latitude and longitude coordinates into plane rectangular coordinates. The specific conversion formula is determined according to the selected projection system.

3. The method for generating a three-dimensional model of an underground pipe network based on data fusion according to claim 1, characterized in that: The depth data acquired by GPR is processed by mean filtering. gpr (i), the filtered value Calculated by the following formula: Where n is an odd-numbered filter window size, which is used to remove data noise.

4. The method for generating a three-dimensional model of an underground pipe network based on data fusion according to claim 1, characterized in that: When constructing a three-dimensional geometric model of a pipeline, for a cylindrical pipeline, the central axis of the pipeline is used as the reference, and according to the pipe diameter D, the coordinates of the pipeline surface points are determined using parametric equations to construct the three-dimensional shape of the pipeline. Parametric equation of the central axis It is related to the pipe diameter and angle θ.

5. The method for generating a three-dimensional model of an underground pipe network based on data fusion according to claim 1, characterized in that: In the data fusion step, the weight is calculated by the data error estimate and the acquisition time. The data with small error and recent acquisition time has higher weight. When calculating the error weight, the GIS data error estimate σ gis and GPR data error estimate σ gpr , calculate the respective error weights according to a specific formula.

6. The method for generating a three-dimensional model of an underground pipe network based on data fusion according to claim 5, characterized in that: The formula for calculating the error weight is:

7. The method for generating a three-dimensional model of an underground pipe network based on data fusion according to claim 5, characterized in that: When calculating the timeliness weight, the time t that has passed since the GIS data was collected is used. gis and the time t that has passed since the GPR data was collected gpr , an exponential decay function is used to determine the timeliness weight, where the decay coefficient λ is adjusted according to the actual situation.

8. The method for generating a three-dimensional model of an underground pipe network based on data fusion according to claim 5, characterized in that: The formula for calculating the timeliness weight is:

9. The method for generating a three-dimensional model of an underground pipe network based on data fusion according to claim 5, characterized in that: The error weight and timeliness weight are combined to obtain the comprehensive weight, and then the attribute value v obtained from the GIS data is calculated by the comprehensive weight. gis And the attribute value v obtained from GPR data gpr Fusion is performed to obtain the fused value v fused .

10. The underground pipe network three-dimensional model generation system based on data fusion is characterized by: Includes the following modules: Multi-source data acquisition module: Use geographic information system GIS to obtain geographical location data of underground pipelines, use pipeline detection robots to collect internal structure data of pipelines, and use ground penetrating radar GPR to measure the depth information of underground pipelines; Data preprocessing module: The module performs coordinate conversion on GIS data, converting the longitude and latitude coordinates into plane rectangular coordinates suitable for modeling; the mean filter method is used to denoise the depth data obtained by GPR to improve data accuracy; Dynamic weight data fusion module: dynamically calculates the weight of GIS and GPR data based on the error estimation and timeliness of data collection; combines the preprocessed GIS coordinate data with the GPR depth data to generate the three-dimensional coordinates of the pipeline network nodes, and constructs the three-dimensional geometric model of the pipeline based on the internal structure data of the pipeline; Model building and visualization module: Utilize the integrated pipe network data and the constructed geometric model, with the help of 3D modeling software or professional GIS 3D modeling module to build a 3D model of the underground pipe network, and perform rendering and visualization operations to intuitively present the distribution and structure of the underground pipe network.

Images