A tunnel comprehensive advanced geological prediction multi-source data three-dimensional fusion imaging method and system

Abstract

The present application relates to a kind of tunnel comprehensive advance geological prediction multi-source data three-dimensional fusion imaging method and system, belong to tunnel engineering and three-dimensional visualization technical field.For the problem of the data format of existing geophysical prospecting method being different, coordinate system being disorder and single data existing multiple solutions, the present application sequentially executes: collecting heterogeneous data such as seismic wave, geological radar, transient electromagnetic etc.;Format analysis and coordinate unification are carried out, and unified three-dimensional object model is constructed;Abnormal value cleaning is carried out;Three-dimensional attribute body model is generated based on adaptive neighborhood constraint and spatial interpolation algorithm;In the unified coordinate system, the fusion risk determination function containing weight distribution and consistency constraint factor is constructed, and the three-dimensional body model of fusion risk is output, so as to reduce multiple solutions;Finally, data thinning and LOD grading control are carried out on the model, and the lightweight three-dimensional display of browser end is realized through visualization rendering pipeline.The present application realizes the spatial consistency management of multi-source data, and significantly improves the accuracy of adverse geological identification and the rendering efficiency of system.

Description

Technical Field

[0001] This invention relates to a method and system for three-dimensional fusion imaging of multi-source data for integrated advanced geological forecasting of tunnels, belonging to the fields of tunnel engineering and three-dimensional visualization technology. Background Technology

[0002] In the tunnel construction risk prevention and control system, the standardized processing of multi-source geological forecast data is the foundation for accurate risk identification. Currently commonly used advanced forecasting methods include seismic wave methods (primarily used for identifying rock strata interfaces and faults), transient electromagnetic methods (primarily used for aquifer detection), and ground-penetrating radar methods (primarily used for identifying shallow adverse geological conditions). In existing technologies, multi-source geophysical data are often stored and analyzed discretely, resulting in the following technical shortcomings:

[0003] Heterogeneous data formats and coordinate systems: Existing seismic wave instruments mostly use the SVY format, ground-penetrating radar uses the DZT format, and transient electromagnetic data uses the DAT format. The local coordinate systems and physical property dimensions differ among the data sources from different instruments, making it impossible to uniformly represent multi-source data in the same three-dimensional space.

[0004] Multiple interpretations lead to inaccurate forecasts: Single geophysical methods are limited by their physical premises and suffer from serious interpretability issues. Existing systems mostly display simple data overlays, lacking in-depth mathematical fusion mechanisms and spatial consistency checks, resulting in blurred boundaries of unfavorable geological conditions.

[0005] Performance bottleneck of 3D visualization rendering: The data generated by advanced geological forecasting is mostly massive point cloud or voxel mesh. Traditional client software is not only cumbersome to operate, but also suffers from browser memory overflow, slow loading, and inability to interact smoothly when the model is ported to the web or mobile terminal for display. Summary of the Invention

[0006] This invention provides a method and system for three-dimensional fusion imaging of multi-source data for integrated advanced geological forecasting of tunnels, in order to solve the problems of difficulty in fusing multi-source heterogeneous data, multiple interpretations of single-source data, and low rendering efficiency on the web.

[0007] The technical solution of this invention is: a three-dimensional fusion imaging method for multi-source data in integrated advanced geological prediction of tunnels, comprising the following steps:

[0008] S1: Collect multi-source advanced geological forecast raw data for the tunnel construction area. The raw data includes at least seismic wave data, ground-penetrating radar data, transient electromagnetic data, and advanced borehole wave velocity data.

[0009] S2: Perform format parsing and physical parameter extraction on each source raw data, and perform unified spatial coordinate transformation and attribute value mapping to construct a unified three-dimensional object model dataset. Each data unit in the unified three-dimensional object model dataset contains global spatial coordinates (X, Y, Z), attribute value H, and data source identification information.

[0010] S3: Perform outlier identification and data cleaning on the unified 3D object model dataset to obtain a standardized 3D point set;

[0011] S4: Within the set three-dimensional space, the standardized three-dimensional point set is divided into regular grids, and a spatial interpolation algorithm is used to generate three-dimensional attribute volume models corresponding to each data source. During the interpolation process, a three-dimensional neighborhood search ellipsoid is set, and the neighborhood radius and the number of sampling points participating in the calculation are dynamically constrained according to the sampling density.

[0012] S5: Perform spatial overlap analysis on the three-dimensional attribute volume models of each data source under a unified coordinate system, construct a fusion risk judgment function at each spatial point P(X,Y,Z) for consistency verification, and output the fusion risk three-dimensional volume model;

[0013] S6: Perform data thinning and hierarchical detail control processing on the fused risk 3D volume model to construct a visualization rendering data pipeline and realize 3D fusion imaging on the terminal.

[0014] Furthermore, in step S2, performing the unified transformation of spatial coordinates and the mapping of attribute values ​​includes:

[0015] Perform a translation matrix transformation from the local engineering coordinate system to the global coordinate system on the seismic wave data, and map the waveform amplitude to the attribute value H according to the gain coefficient;

[0016] The Z-coordinate is obtained by performing a time-depth transformation on the ground-penetrating radar data, and the waveform features are extracted as the attribute value H.

[0017] The transient electromagnetic data is subjected to voltage-resistivity conversion and deep inversion calculation to obtain the Z coordinate, with resistivity as the attribute value H.

[0018] Perform three-dimensional coordinate mapping of the borehole trajectory on the borehole wave velocity data, with wave velocity as the attribute value H.

[0019] Furthermore, in step S4, the spatial interpolation algorithm is at least one of inverse distance weighted interpolation, local polynomial interpolation, or Kriging interpolation.

[0020] Furthermore, when using the inverse distance weighted interpolation method, the distance power exponent n is 2, and the search radius of the three-dimensional neighborhood search ellipsoid is independently set in the X, Y, and Z directions.

[0021] Furthermore, in S4, the constraint on the number of sampling points includes:

[0022] Set the minimum and maximum number of sampling points. When the number of data points in the neighborhood exceeds the maximum number of sampling points, sort them by spatial distance from nearest to farthest, and select the preset number of data points that are closest to the nearest point to participate in the interpolation calculation.

[0023] Furthermore, in step S4, the attribute values ​​of each data source are first standardized to obtain standardized attribute values. ;

[0024] Constructing the basic risk weighting function: ,in Here are the weight coefficients for each data source, and , where n is the total number of data sources.

[0025] Furthermore, in step S5, the introduction of a consistency constraint factor is also included. The calculation formula is as follows:

[0026] ;

[0027] in, This represents the number of data sources whose attribute values ​​at point P meet the corresponding anomaly threshold. This represents the total number of data sources participating in the evaluation.

[0028] Furthermore, the final fusion risk value at point P Calculated using the following formula:

[0029] ;

[0030] when If the risk level exceeds the set comprehensive risk threshold, the spatial point P is determined to belong to an area with adverse geological risks.

[0031] Furthermore, the adverse geological bodies characterized by the fusion risk three-dimensional volume model include low-resistivity anomalies, water-bearing anomalies, or fault fracture zones.

[0032] Furthermore, the data thinning and hierarchical detail control processing described in step S6 includes: point cloud thinning based on spatial sampling rate, and LOD (Level of Detail) hierarchical mesh loading based on visual distance.

[0033] Furthermore, in step S6, the visualization rendering data pipeline is constructed using WebGL and VTK.js technologies. The rendering pipeline includes a data conversion module, a filtering module, a graphics processing module, and a drawing module in sequence.

[0034] Furthermore, the terminal implements 3D fusion imaging that supports volume rendering, surface rendering, contour extraction, dynamic threshold adjustment, and spherical cropping operations.

[0035] This invention also provides a three-dimensional fusion imaging system for multi-source data in tunnel integrated advanced geological prediction, comprising:

[0036] The data acquisition module is used to collect raw data from multiple sources of advanced geological forecasting in the tunnel construction area;

[0037] The standardization parsing module is used to perform format parsing, coordinate transformation and attribute value mapping on the original data from various sources, and to construct a unified three-dimensional object model dataset containing (X,Y,Z,H).

[0038] The interpolation modeling module is used to generate a 3D attribute volume model corresponding to each data source by applying a spatial interpolation algorithm with adaptive neighborhood constraints to the cleaned 3D point set.

[0039] The fusion judgment module is used to calculate the final fusion risk value of each point in space under a unified coordinate system by using the fusion risk judgment function and consistency constraint factor, and output the fusion risk three-dimensional volume model.

[0040] The lightweight rendering module is used to perform data thinning and LOD hierarchical control on the fusion risk 3D volume model, and realize 3D fusion imaging on the terminal.

[0041] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels.

[0042] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels.

[0043] The beneficial effects of this invention are:

[0044] 1. A unified object model standard was established: By extracting (X,Y,Z) coordinates and attribute values ​​H, the barriers of special formats such as SVY, DZT, and DAT were broken, and standardized management of multi-source data under the same spatial architecture was realized.

[0045] 2. Significantly reduces the ambiguity of geological forecasts: A mathematical fusion model incorporating weighted allocation R(P) and consistency constraint factor C(P) is creatively proposed. Only when multiple detection methods simultaneously point to anomalies (high consistency factor) will the final risk value be significantly amplified, thereby effectively filtering out false anomalies generated by single-source equipment and improving the accuracy of forecasts;

[0046] 3. Balancing modeling accuracy and computational efficiency: Adaptive ellipsoidal neighborhood constraints and maximum / minimum sampling point constraints are introduced in spatial interpolation to avoid the waste of computing power caused by global interpolation of massive discrete data;

[0047] 4. Achieved lightweight, cross-platform 3D imaging: Through data thinning, LOD grading, and custom rendering pipeline based on VTK.js, voxel rendering, which originally required heavy graphics workstations, can be executed smoothly in ordinary browsers and mobile devices, truly empowering front-line construction safety management. Attached Figure Description

[0048] Figure 1 This is a flowchart of the multi-source data acquisition and standardized analysis system of the present invention;

[0049] Figure 2 This is a schematic diagram illustrating the conversion of multi-source data from its original format to a unified (X,Y,Z,H) object model.

[0050] Figure 3 Algorithm logic diagram for outputting a risk volume model through multi-source 3D volume fusion;

[0051] Figure 4 This is a schematic diagram of the lightweight rendering and browser-side display structure based on VTK.js;

[0052] Figure 5 This is a result of three-dimensional fusion imaging of multi-source data for comprehensive advanced geological forecasting of a tunnel. Detailed Implementation

[0053] The technical solution of the present invention will be clearly and completely described below with reference to specific embodiments;

[0054] Example 1: As Figures 1-5 As shown, a three-dimensional fusion imaging method for multi-source data in tunnel integrated advanced geological prediction includes the following steps:

[0055] S1: Collect multi-source advanced geological forecast raw data for the tunnel construction area. The raw data includes at least seismic wave data, ground-penetrating radar data, transient electromagnetic data, and advanced borehole wave velocity data.

[0056] S2: Perform format parsing and physical parameter extraction on each source raw data, and perform unified spatial coordinate transformation and attribute value mapping to construct a unified three-dimensional object model dataset. Each data unit in the unified three-dimensional object model dataset contains global spatial coordinates (X, Y, Z), attribute value H, and data source identification information.

[0057] Furthermore, in step S2, performing the unified transformation of spatial coordinates and the mapping of attribute values ​​includes:

[0058] Perform a translation matrix transformation from the local engineering coordinate system to the global coordinate system on the seismic wave data, and map the waveform amplitude to the attribute value H according to the gain coefficient;

[0059] The Z-coordinate is obtained by performing a time-depth transformation on the ground-penetrating radar data, and the waveform features are extracted as the attribute value H.

[0060] The transient electromagnetic data is subjected to voltage-resistivity conversion and deep inversion calculation to obtain the Z coordinate, with resistivity as the attribute value H.

[0061] Perform three-dimensional coordinate mapping of the borehole trajectory on the borehole wave velocity data, with wave velocity as the attribute value H.

[0062] S3: Perform outlier identification and data cleaning on the unified 3D object model dataset to obtain a standardized 3D point set;

[0063] S4: Within the set three-dimensional space, the standardized three-dimensional point set is divided into regular grids, and a spatial interpolation algorithm is used to generate three-dimensional attribute volume models corresponding to each data source. During the interpolation process, a three-dimensional neighborhood search ellipsoid is set, and the neighborhood radius and the number of sampling points participating in the calculation are dynamically constrained according to the sampling density.

[0064] Furthermore, in step S4, the spatial interpolation algorithm is at least one of inverse distance weighted interpolation, local polynomial interpolation, or Kriging interpolation.

[0065] Furthermore, when using the inverse distance weighted interpolation method, the distance power exponent n is 2, and the search radius of the three-dimensional neighborhood search ellipsoid is independently set in the X, Y, and Z directions.

[0066] Furthermore, in S4, the constraint on the number of sampling points includes:

[0067] Set the minimum and maximum number of sampling points. When the number of data points in the neighborhood exceeds the maximum number of sampling points, sort them by spatial distance from nearest to farthest, and select the preset number of data points that are closest to the nearest point to participate in the interpolation calculation.

[0068] Furthermore, in step S4, the attribute values ​​of each data source are first standardized to obtain standardized attribute values. ;

[0069] Constructing the basic risk weighting function: ,in Here are the weight coefficients for each data source, and , where n is the total number of data sources.

[0070] Furthermore, in step S5, the introduction of a consistency constraint factor is also included. The calculation formula is as follows:

[0071] ;

[0072] in, This represents the number of data sources whose attribute values ​​at point P meet the corresponding anomaly threshold. This represents the total number of data sources participating in the evaluation.

[0073] Furthermore, the final fusion risk value at point P Calculated using the following formula:

[0074] ;

[0075] when If the risk level exceeds the set comprehensive risk threshold, the spatial point P is determined to belong to an area with adverse geological risks.

[0076] This invention also provides a three-dimensional fusion imaging system for multi-source data in tunnel integrated advanced geological prediction, comprising:

[0077] The data acquisition module is used to collect raw data from multiple sources of advanced geological forecasting in the tunnel construction area;

[0078] The standardization parsing module is used to perform format parsing, coordinate transformation and attribute value mapping on the original data from various sources, and to construct a unified three-dimensional object model dataset containing (X,Y,Z,H).

[0079] The interpolation modeling module is used to generate a 3D attribute volume model corresponding to each data source by applying a spatial interpolation algorithm with adaptive neighborhood constraints to the cleaned 3D point set.

[0080] The fusion judgment module is used to calculate the final fusion risk value of each point in space under a unified coordinate system by using the fusion risk judgment function and consistency constraint factor, and output the fusion risk three-dimensional volume model.

[0081] The lightweight rendering module is used to perform data thinning and LOD hierarchical control on the fusion risk 3D volume model, and realize 3D fusion imaging on the terminal.

[0082] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels.

[0083] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels.

[0084] Example 2: A three-dimensional fusion imaging method for multi-source data in tunnel integrated advanced geological prediction, comprising the following steps:

[0085] Step 1: In a certain tunnel project, raw data of seismic waves (SVY format), ground-penetrating radar (DZT format), and transient electromagnetic data (DAT format) were acquired on site.

[0086] Step 2: First, a unified 3D object model transformation is performed: For seismic wave data, the local coordinates of the excitation point and sensor are read and converted into global coordinates using the coordinate offset matrix (x+ΔX,y+ΔY,z+ΔZ), and the amplitude is converted into the attribute value H; for ground-penetrating radar data, the reflection time t is converted into depth Z according to the electromagnetic wave velocity v=c / ϵr; for transient electromagnetic data, the induced voltage is extracted and converted into depth Z and resistivity attribute value H through apparent resistivity inversion calculation; finally, all heterogeneous data are formatted into the standard format P={X,Y,Z,H,SourceType}.

[0087] Step 3: Perform outlier identification and data cleaning on the unified 3D object model dataset to obtain a standardized 3D point set;

[0088] Step 4: After data cleaning, the discrete points need to be meshed. The system provides Inverse Distance Weighted Interpolation (IDW) and Local Polynomial Interpolation. Taking IDW as an example, to improve the calculation speed and take into account the characteristics of the tunnel's narrow and elongated orientation, an anisotropic search ellipsoid (i.e., different search radii for the X, Y, and Z axes) is set. The system dynamically limits the number of calculation points to Nmin to Nmax. When the number of neighboring points is large, only the nearest Nmax points are selected for weight allocation. The weight factor is defined as one-squared of the distance from the i-th neighboring sampling point to the current interpolation position, thereby significantly reducing the computational complexity and generating a high-precision three-dimensional attribute volume model.

[0089] Step 5: To address the issue of multiple solutions, this embodiment performs multi-source fusion at spatial grid point P(X,Y,Z) and outputs a 3D volumetric model of the fusion risk.

[0090] First, physical quantities such as radar reflection intensity and inverted resistivity are normalized to dimensionless values. ; Calculate the underlying risk: ;

[0091] Subsequently, the number of methods that reached their respective abnormal thresholds at that point was counted. If two out of the three methods show abnormalities, then the consistency factor... ;

[0092] Final risk value This algorithm mechanism ensures that noise data from isolated devices is suppressed, and the real risks of multi-source resonance (such as water-rich fracture zones) are highlighted.

[0093] Step 6: After obtaining the fusion risk model, due to the data volume reaching GB levels, the system employs a thinning algorithm and LOD technology for processing, and stores it through distributed object storage OSS (MinIO cluster). In the front-end application, a visualization pipeline is built based on VTK.js. The data sequentially passes through Source loading, Filter filtering (such as spherical truncation calculation), Mapper mapping, and finally, the Actor performs volume rendering or iso-surface rendering on the WebGL canvas. Frontline construction personnel can not only intuitively browse the 3D geological body in front of the tunnel through mobile APP or Web, but also adjust the threshold in real time by dragging the slider to strip away normal surrounding rock and expose high-risk adverse geological bodies, thereby providing one-click guidance for the construction progress at the tunnel face.

[0094] Figure 5 This is a 3D fusion imaging result of multi-source data from a tunnel comprehensive advanced geological prediction system, capable of slicing at any location on the 3D model. The black geological bodies represent karst cavities extracted from ground-penetrating radar data in the tunnel's advanced direction. The yellow and green backgrounds represent the velocity attributes of the TSP data, with yellow indicating high velocity and green indicating low velocity, and green also indicating fractured surrounding rock or high weathering. The blue areas represent transient electromagnetic resistivity data, with blue indicating low resistivity, which, combined with geological conditions, suggests potential water-rich locations. The fusion of three geophysical methods into a 3D image allows for a comprehensive assessment of geological risks in the tunnel face, both in the advanced and reverse directions.

[0095] The specific embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A three-dimensional fusion imaging method for multi-source data in integrated advanced geological prediction of tunnels, characterized in that: Includes the following steps: S1: Collect multi-source advanced geological forecast raw data for the tunnel construction area. The raw data includes at least seismic wave data, ground-penetrating radar data, transient electromagnetic data, and advanced borehole wave velocity data. S2: Perform format parsing and physical parameter extraction on each source raw data, and perform unified spatial coordinate transformation and attribute value mapping to construct a unified three-dimensional object model dataset. Each data unit in the unified three-dimensional object model dataset contains global spatial coordinates (X, Y, Z), attribute value H, and data source identification information. S3: Perform outlier identification and data cleaning on the unified 3D object model dataset to obtain a standardized 3D point set; S4: Within the set three-dimensional space, the standardized three-dimensional point set is divided into regular grids, and a spatial interpolation algorithm is used to generate three-dimensional attribute volume models corresponding to each data source. During the interpolation process, a three-dimensional neighborhood search ellipsoid is set, and the neighborhood radius and the number of sampling points participating in the calculation are dynamically constrained according to the sampling density. S5: Perform spatial overlap analysis on the three-dimensional attribute volume models of each data source under a unified coordinate system, construct a fusion risk judgment function at each spatial point P(X,Y,Z) for consistency verification, and output the fusion risk three-dimensional volume model; S6: Perform data thinning and hierarchical detail control processing on the fused risk 3D volume model to construct a visualization rendering data pipeline and realize 3D fusion imaging on the terminal.

2. The method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels according to claim 1, characterized in that: In step S2, the unified transformation of spatial coordinates and mapping of attribute values ​​include: Perform a translation matrix transformation from the local engineering coordinate system to the global coordinate system on the seismic wave data, and map the waveform amplitude to the attribute value H according to the gain coefficient; The Z-coordinate is obtained by performing a time-depth transformation on the ground-penetrating radar data, and the waveform features are extracted as the attribute value H. The transient electromagnetic data is subjected to voltage-resistivity conversion and deep inversion calculation to obtain the Z coordinate, with resistivity as the attribute value H. Perform three-dimensional coordinate mapping of the borehole trajectory on the borehole wave velocity data, with wave velocity as the attribute value H.

3. The method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels according to claim 1, characterized in that: In step S4, the spatial interpolation algorithm is at least one of inverse distance weighted interpolation, local polynomial interpolation, or Kriging interpolation.

4. The method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels according to claim 3, characterized in that: When the inverse distance weighted interpolation method is used, the distance power exponent n is 2, and the search radius of the three-dimensional neighborhood search ellipsoid is set independently in the X, Y, and Z directions.

5. The method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels according to claim 1, characterized in that: In S4, the sampling point number constraint includes: Set the minimum and maximum number of sampling points. When the number of data points in the neighborhood exceeds the maximum number of sampling points, sort them by spatial distance from nearest to farthest, and select the preset number of data points that are closest to the nearest point to participate in the interpolation calculation.

6. The method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels according to claim 1, characterized in that: In step S4, the attribute values ​​of each data source are first standardized to obtain standardized attribute values. ; Constructing the basic risk weighting function: ,in Here are the weight coefficients for each data source, and , where n is the total number of data sources.

7. The method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels according to claim 6, characterized in that: In step S5, the process further includes introducing a consistency constraint factor. The calculation formula is as follows: ； in, This represents the number of data sources whose attribute values ​​at point P meet the corresponding anomaly threshold. This represents the total number of data sources participating in the evaluation.

8. The method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels according to claim 6, characterized in that: Final fusion risk value at point P Calculated using the following formula: ； when If the risk level exceeds the set comprehensive risk threshold, the spatial point P is determined to belong to an area with adverse geological risks.

9. The method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels according to claim 1, characterized in that: The adverse geological bodies characterized by the fusion risk 3D volume model include low-resistivity anomalies, water-bearing anomalies, or fault fracture zones.

10. The method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels according to claim 1, characterized in that: The data thinning and hierarchical detail control processing in step S6 includes: point cloud thinning based on spatial sampling rate, and LOD (Level of Detail) hierarchical mesh loading based on visual distance.

11. The method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels according to claim 1, characterized in that: In step S6, the visualization rendering data pipeline is constructed using WebGL and VTK.js technologies. The rendering pipeline includes a data conversion module, a filtering module, a graphics processing module, and a drawing module in sequence.

12. The method for three-dimensional fusion imaging of multi-source data for integrated advanced geological prediction of tunnels according to claim 1, characterized in that: The terminal enables 3D fusion imaging that supports volume rendering, surface rendering, contour extraction, dynamic threshold adjustment, and spherical cropping.

13. A three-dimensional fusion imaging system for multi-source data in tunnel integrated advanced geological prediction, characterized in that, include: The data acquisition module is used to collect raw data from multiple sources of advanced geological forecasting in the tunnel construction area; The standardization parsing module is used to perform format parsing, coordinate transformation and attribute value mapping on the original data from various sources, and to construct a unified three-dimensional object model dataset containing (X,Y,Z,H). The interpolation modeling module is used to generate a 3D attribute volume model corresponding to each data source by applying a spatial interpolation algorithm with adaptive neighborhood constraints to the cleaned 3D point set. The fusion judgment module is used to calculate the final fusion risk value of each point in space under a unified coordinate system by using the fusion risk judgment function and consistency constraint factor, and output the fusion risk three-dimensional volume model. The lightweight rendering module is used to perform data thinning and LOD hierarchical control on the fusion risk 3D volume model, and realize 3D fusion imaging on the terminal.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the three-dimensional fusion imaging method for multi-source data in tunnel integrated advanced geological prediction as described in any one of claims 1 to 12.

15. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the three-dimensional fusion imaging method for multi-source data in tunnel integrated advanced geological prediction as described in any one of claims 1 to 12.

Images