Rapid karst cave modeling method and system based on multi-source data

Abstract

The invention provides a rapid karst cave modeling method and system based on multi-source data, and the method comprises the steps: collecting multi-source data including geological exploration data, remote sensing data and unmanned aerial vehicle aerial photography data, and carrying out the preprocessing of the multi-source data; fusing the preprocessed data by adopting a weighted fusion or principal component analysis method; extracting shape features, size features and position features from the fused data; the extracted karst cave features are converted into point cloud data or grid data, and a karst cave model is constructed through three-dimensional modeling software; performing vertex editing and patch adjustment on the initial model, and correcting the model form in combination with expert experience; and comparing the constructed karst cave model with actual geological data, and if an error exceeds a preset threshold value, iteratively adjusting model parameters until a precision requirement is met. According to the method, efficient extraction, accurate modeling and dynamic optimization of the karst cave geological information can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of geological modeling, and in particular to a method and system for rapid cave modeling based on multi-source data, which aims to realize rapid and accurate construction of a cave model by integrating multiple data sources. Background Art

[0002] In the current field of cave geological modeling, with the continuous advancement of information technology, data collection methods are becoming increasingly rich, including geological exploration data, remote sensing images, drone aerial photography and other data sources. However, these data sources are often independent and lack unified data processing standards and fusion mechanisms, resulting in low data utilization efficiency and difficulty in fully and accurately reflecting the true geological characteristics of the cave. In addition, most of the existing cave modeling methods rely on a single data source, the modeling process is cumbersome and the accuracy is limited, which cannot meet the needs of fast and accurate modeling. Therefore, how to efficiently integrate multi-source data to achieve accurate extraction of cave characteristics and rapid construction of three-dimensional models has become a key issue that needs to be urgently solved in the current field of cave modeling.

[0003] The main problems with current cave modeling technology are insufficient data fusion, inaccurate feature extraction, and a lack of model verification and optimization methods. First, the format differences and inconsistent coordinate systems between multi-source data limit the effective fusion of data, resulting in the incomplete and inaccurate basic data for model construction. Secondly, the cave feature extraction method is relatively backward, and it is difficult to accurately identify the key features of the cave from complex geological data, which affects the accuracy and practicality of the model. Finally, the model verification and optimization steps lack systematicity and scientificity, and often rely on manual experience and judgment, making it difficult to ensure the accuracy and reliability of the model. These problems jointly restrict the development and application of cave modeling technology. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for rapid cave modeling based on multi-source data, which can realize efficient extraction, accurate modeling and dynamic optimization of cave geological information through multi-source data fusion, intelligent analysis and three-dimensional modeling technology, thereby providing high-precision and high-reliability cave geological model support for geological research, resource exploration, disaster prevention and other fields.

[0005] A method for rapid cave modeling based on multi-source data includes the following steps:

[0006] Step S101: collecting multi-source data including geological exploration data, remote sensing data and drone aerial photography data, and pre-processing the multi-source data;

[0007] Step S102: The pre-processed data is fused using weighted fusion or principal component analysis (PCA) to form a comprehensive cave geological information dataset:

[0008] Step S103: Extract shape features, size features, and location features from the fused data using machine learning algorithms or deep learning models;

[0009] Step S104: Convert the extracted karst cave features into point cloud data or mesh data, and use 3D modeling software to construct a karst cave model; perform vertex editing and patch adjustment on the initial model through a human-computer interaction tool, and correct the model morphology in combination with expert experience;

[0010] Step S105: Compare the constructed karst cave model with the actual geological data, and calculate the spatial error, morphological error, and volume error; if the error exceeds the preset threshold, iteratively adjust the model parameters until the accuracy requirement is met.

[0011] Furthermore, the geological exploration data includes borehole data, seismic data, and geological profiles, the remote sensing data includes satellite images and radar data, and the UAV aerial survey data includes LiDAR point clouds and high-resolution images.

[0012] Furthermore, the preprocessing of the multi-source data includes:

[0013] Data cleaning: Remove noise through filtering methods, interpolate and fill missing data, and correct incorrect data;

[0014] Format unification: The format unification includes format conversion and coordinate system unification. Format conversion is to convert geological exploration data, remote sensing data, and UAV aerial survey data into geospatial raster vector formats;

[0015] Data calibration: Based on ground control points, spatially align the multi-source data to eliminate spatial deviations caused by sensors or platforms; align the time series of the multi-source data through spectral analysis to ensure the consistency of the data on the time axis.

[0016] Furthermore, the weighted fusion includes dynamically allocating weights according to the accuracy and resolution of the data sources to generate a comprehensive data set;

[0017] The PCA fusion includes extracting principal components through covariance matrix decomposition, eliminating redundant information, and constructing a dimensionality-reduced data set.

[0018] Furthermore, Step S104 specifically includes:

[0019] Step S501: By integrating, preprocessing, and extracting features from multi-source data, a high-precision karst cave model is constructed using 3D modeling technology. The extracted karst cave feature dataset includes shape features, size features, and position features. Shape features include the major axis, minor axis, and orientation angle of the karst cave. Size features include the diameter or volume of the karst cave. Position features include the longitude and latitude coordinates or relative position of the karst cave. The feature data is converted into point cloud data or mesh data supported by 3D modeling software.

[0020] Among them, point cloud data is generated according to the position and shape features of the karst cave. Assuming the center coordinates of the karst cave are (x c , y c , z c ), the major axis is a, the minor axis is b, and the orientation angle is θ. Then the calculation formula for the point cloud coordinates (x, y, z) is:

[0021] x = x c + a·cos(φ)·cos(θ) - b·sin(φ)·sin(θ)

[0022] y = y c + a·cos(φ)·cos(θ) + b·sin(φ)·sin(θ)

[0023] z = z c + c·sin(φ)

[0024] Among them, φ is the parameter angle, 0 ≤ φ < 2π, and c is the height of the karst cave.

[0025] The point cloud data is converted into mesh data, and triangular patches are generated using the triangulation algorithm. The formula for triangulation is:

[0026]

[0027] Among them, v1, v2, v3 are the vertex coordinates of the triangular patch.

[0028] The generated mesh data is imported into 3D modeling software, and the 3D model is rendered in the software, and the material and lighting effects are set.

[0029] Step S502: Through the visualization interface and interaction tools, combined with expert experience and user input, the model is dynamically adjusted and optimized. The specific steps are as follows:

[0030] (1) Interactive interface design

[0031] The interface provides basic operations such as translation, rotation, and scaling, as well as advanced functions such as vertex editing and patch adjustment, and interacts with the model through the mouse, keyboard, or touch device.

[0032] (2) Model adjustment

[0033] Vertex editing: Select the vertices in the model and adjust their positions to optimize the model shape. Assume the original coordinates of vertex vi are (x i , y i , z i ), and the adjusted coordinates are (x i ′, y i ′, z i ). Then the vertex displacement vector is:

[0034] Δv i =(x i ′ - x i , y i ′ - y i , z i ′ - z i )

[0035] Triangle patch adjustment: Select the triangular patches in the model and adjust their normal directions or areas. Assume the normal vector of the triangular patch is n, and the adjusted normal vector is n′. Then the normal rotation matrix is:

[0036]

[0037] where θ is the rotation angle;

[0038] (3) Real-time preview

[0039] During the adjustment process, preview the model changes in real time to ensure that the adjustment effect meets the expectations. Use the level of detail technology to dynamically adjust the model details according to the viewing distance;

[0040] Step S503: Continuously optimize the model accuracy by iteratively adjusting the model parameters, comparing with the actual geological data, and using the cross-validation method, and ensure the high accuracy and reliability of the model through error analysis and reliability evaluation. The cave model parameters include shape, size, volume, topological structure, physical properties, the number of vertices and polygons, and position. The specific steps are as follows:

[0041] (1) Model optimization

[0042] Use the Laplace smoothing algorithm to smooth the model surface. Assume the set of neighbor vertices of vertex v i is N(i). Then the smoothed vertex coordinates are:

[0043]

[0044] Use the edge collapse algorithm to simplify the model and reduce the number of triangular patches. Assume the edge e ij connects vertices v i and v j . The coordinates of the new vertex after folding are:

[0045]

[0046] (2), Model Verification

[0047] Check whether the geometric properties of the model are consistent with the original data. Assume the volume of the model is V and the surface area is S, then the calculation formula is:

[0048]

[0049] Among them, T is the triangular facet. Visually compare the optimized model with the original data to check the accuracy and authenticity of the model;

[0050] (3), Result Output

[0051] Export the optimized model into common formats, and the common formats include OBJ, STL, and FBX.

[0052] A rapid karst cave modeling system based on multi-source data, comprising:

[0053] A multi-source data acquisition and preprocessing module, configured to collect multi-source data including geological exploration data, remote sensing data, and UAV aerial photography data, and preprocess the multi-source data;

[0054] A multi-source data fusion module, configured to fuse the preprocessed data by using a weighted fusion or principal component analysis (PCA) method to form a comprehensive karst cave geological information dataset:

[0055] A karst cave feature extraction module, configured to extract shape features, size features, and position features from the fused data by using machine learning algorithms or deep learning models;

[0056] A 3D modeling and interactive optimization module, configured to convert the extracted karst cave features into point cloud data or mesh data, and construct a karst cave model by using 3D modeling software; perform vertex editing and facet adjustment on the initial model through a human-computer interaction tool, and correct the model shape in combination with expert experience;

[0057] A model verification and iterative optimization module, configured to compare the constructed karst cave model with actual geological data, calculate spatial error, morphological error, and volume error; if the error exceeds a preset threshold, iteratively adjust the model parameters until the accuracy requirement is met.

[0058] Further, the geological exploration data includes borehole data, seismic data, and geological profiles, the remote sensing data includes satellite images and radar data, and the UAV aerial photography data includes LiDAR point clouds and high-resolution images.

[0059] Further, the multi-source data acquisition and preprocessing module preprocesses the multi-source data, specifically including:

[0060] Data cleaning: Remove noise through filtering methods, interpolate and fill in missing data, and correct incorrect data;

[0061] Format unification: Format unification includes format conversion and coordinate system unification;

[0062] Data calibration: Based on ground control point pairs, spatially align the multi-source data to eliminate spatial deviations caused by sensors or platforms; Align the time series of the multi-source data through spectral analysis to ensure the consistency of the data on the time axis;

[0063] Format conversion: Convert geological exploration data, remote sensing data, and LiDAR point cloud data into geospatial raster vector formats.

[0064] Further, the multi-source data fusion module performs weighted fusion on the preprocessed data, including dynamically assigning weights according to the accuracy and resolution of the data sources to generate a comprehensive data set; The multi-source data fusion module performs fusion on the preprocessed data using the principal component analysis (PCA) method, including extracting principal components through covariance matrix decomposition, eliminating redundant information, and constructing a dimensionality-reduced data set.

[0065] Further, the three-dimensional modeling and interaction optimization module is specifically used for,

[0066] By integrating, preprocessing, and extracting features from multi-source data, use three-dimensional modeling technology to construct a high-precision karst cave model, where the extracted karst cave feature data set includes shape features, size features, and location features; Shape features include the major axis, minor axis, and orientation angle of the karst cave; Size features include the diameter or volume of the karst cave; Location features include the longitude and latitude coordinates or relative positions of the karst cave; Convert the feature data into point cloud data or mesh data supported by three-dimensional modeling software;

[0067] Among them, generate point cloud data according to the location and shape features of the karst cave. Assume that the center coordinates of the karst cave are (x c ,y c ,z c ), the major axis is a, the minor axis is b, and the orientation angle is θ. Then the calculation formula for the point cloud coordinates (x, y, z) is:

[0068] x = x c + a·cos(φ)·cos(θ) - b·sin(φ)·sin(θ)

[0069] y = y c + a·cos(φ)·cos(θ) + b·sin(φ)·sin(θ)

[0070] z = z c + c·sin(φ)

[0071] where φ is the parameter angle, 0 ≤ φ < 2π, and c is the height of the karst cave;

[0072] Convert the point cloud data into mesh data and generate triangular patches using the triangulation algorithm. The formula for triangulation is:

[0073]

[0074] where v1, v2, and v3 are the vertex coordinates of the triangular patch.

[0075] Import the generated mesh data into 3D modeling software, render the 3D model in the software, and set the material and lighting effects;

[0076] Through the visualization interface and interaction tools, combine expert experience with user input to dynamically adjust and optimize the model. The specific steps are as follows:

[0077] (1) Interactive interface design

[0078] The interface provides basic operations such as translation, rotation, and scaling, as well as advanced functions such as vertex editing and patch adjustment, and interacts with the model through the mouse, keyboard, or touch device;

[0079] (2) Model adjustment

[0080] Vertex editing: Select the vertices in the model and adjust their positions to optimize the model shape. Assume the original coordinates of vertex vi are (x i , y i , z i ), and the adjusted coordinates are (x i ′, y i ′, z i ). Then the vertex displacement vector is:

[0081] Δv i = (x i ′ - x i , y i ′ - y i , z i ′ - z i )

[0082] Patch adjustment: Select the triangular patches in the model and adjust their normal directions or areas. Assume the normal vector of the triangular patch is n, and the adjusted normal vector is n′. Then the normal rotation matrix is:

[0083]

[0084] where θ is the rotation angle;

[0085] (3), Real-time preview

[0086] During the adjustment process, the model changes are previewed in real time to ensure that the adjustment effect meets the expectations. The level-of-detail technology is used to dynamically adjust the model details according to the viewing distance;

[0087] By iteratively adjusting the model parameters, comparing the actual geological data, and using the cross-validation method, the model accuracy is continuously optimized, and the high accuracy and reliability of the model are ensured through error analysis and reliability evaluation. The parameters of the karst cave model include shape, size, volume, topological structure, physical properties, the number of vertices and polygons, and position. The specific steps are as follows:

[0088] (1), Model optimization

[0089] Use the Laplace smoothing algorithm to smooth the model surface. Assume that the set of neighbor vertices of vertex v i is N(i), then the smoothed vertex coordinates are:

[0090]

[0091] Use the edge collapse algorithm to simplify the model and reduce the number of triangular patches. Assume that edge e ij connects vertex v i and v j , and the coordinates of the new vertex after folding are:

[0092]

[0093] (2), Model verification

[0094] Check whether the geometric properties of the model are consistent with the original data. Assume that the volume of the model is V and the surface area is S, then the calculation formula is:

[0095]

[0096] where T is the triangular patch. The optimized model is visually compared with the original data to check the accuracy and authenticity of the model;

[0097] (3), Result output

[0098] Export the optimized model into common formats, and the common formats include OBJ, STL, and FBX.

[0099] The present invention has the following beneficial effects:

[0100] 1. The present invention realizes the efficient fusion of multi-source data, the accurate extraction of karst cave features, and the rapid construction, verification, and optimization of the 3D model;

[0101] 2. By adopting advanced data preprocessing techniques, the present invention solves the problems of format differences and inconsistent coordinate systems among multi-source data, and realizes seamless data fusion;

[0102] 3. By using machine learning algorithms and deep learning models, the present invention can accurately extract the key features of karst caves from complex geological data, improving the accuracy and practicality of the model;

[0103] 4. The present invention verifies and compares with actual geological data, continuously optimizes and adjusts the model parameters, and ensures the accuracy and reliability of the model.

[0104] The above advantages together constitute the core competitiveness of the present invention in the field of karst cave modeling, providing strong technical support for fields such as karst cave geological exploration and mineral resource development. BRIEF DESCRIPTION OF THE DRAWINGS

[0105] Figure 1 is a flowchart of a rapid karst cave modeling method based on multi-source data according to an embodiment of the present invention;

[0106] Figure 2 is a flowchart of multi-source data collection and preprocessing in an embodiment of the present invention;

[0107] Figure 3 is a flowchart of three-dimensional modeling and interactive optimization of a karst cave model in an embodiment of the present invention;

[0108] Figure 4 is a flowchart of model verification and iterative optimization in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0109] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0110] Please refer to Figures 1-4 , an embodiment of the present invention provides a rapid karst cave modeling method based on multi-source data, including the following steps:

[0111] Step S101: Multi-source data collection and preprocessing

[0112] Collect multi-source data such as geological exploration data, remote sensing data, and UAV aerial photography data, and perform preprocessing, which includes data cleaning, format unification, data calibration, etc., to ensure the accuracy and consistency of the data. Specifically, step S101 includes the following steps:

[0113] Step S201: Through multi-source data collection, comprehensively obtain the spatial distribution, morphological characteristics, and geological attribute information of karst caves, providing reliable data for the accurate construction and verification of the model.

[0114] Specifically, obtain geological exploration data such as geological exploration reports, borehole data, seismic data, and gravity measurement data through on-site exploration, laboratory analysis, historical data collation, etc. Obtain remote sensing data such as satellite images, aerial images, and radar data through satellite receiving stations, aerial photography, radar scanning, etc. Equipment such as high-resolution cameras and lidar (LiDAR) carried by unmanned aerial vehicles obtain UAV aerial photography data such as LiDAR point clouds and high-resolution images through UAV flight missions.

[0115] Step S202: Clean the multi-source data. Specifically, remove noise through filtering methods, interpolate and fill in missing data, and correct incorrect data to ensure the integrity, consistency, and accuracy of the data, providing high-quality data for subsequent model construction and optimization;

[0116] Among them, remove noise from geological exploration data, remote sensing data, and UAV aerial photography data. Using filtering processing technology, adopt the Gaussian filtering method, and the calculation formula is as follows:

[0117]

[0118] In the formula: G(x,y) is the weight value of the Gaussian filter at the point (x,y); x, y are the coordinates of the current pixel point (relative to the center of the filter); σ is the standard deviation of the Gaussian distribution, controlling the smoothness of the filter. The larger σ is, the smoother the filter. e is the natural constant, approximately equal to 2.71828. π is the pi, approximately equal to 3.14159. For a small amount of data, the median filtering method can also be used. For each pixel point, take the median value in its neighborhood as the value of this point.

[0119] Interpolate the missing values of geological exploration data, remote sensing data, and UAV aerial photography data. The linear interpolation formula is as follows:

[0120]

[0121] In the formula: y is the value of the interpolation point (the value to be calculated); x is the position of the interpolation point; x1, x2 are the positions of the known data points; y1, y2 are the values of the known data points. Adopt the Kriging interpolation method, based on spatial autocorrelation, and use the variogram for interpolation.

[0122] Correct the incorrect geological exploration data, remote sensing data, and UAV aerial photography data. The least squares method formula is as follows:

[0123]

[0124] In the formula: y i is the actual observed value. y pi is the model predicted value. N is the total number of data points. The ∑ summation symbol represents minimizing the sum of the squared errors for all data points.

[0125] Step S203: Ensure the consistency and comparability of the data by unifying the formats of multi-source data (such as format conversion, coordinate system unification) and data calibration (such as time synchronization, spatial alignment, and unit conversion).

[0126] The format conversion is as follows: Geological exploration data usually includes borehole data (point data), geological profiles (line data), and geological boundaries (surface data), and the formats may be Excel, CSV, or CAD files. Remote sensing data mainly includes satellite images and aerial images, and the formats are mostly GeoTIFF, JPEG, or HDF. UAV aerial survey data includes high-resolution images and LiDAR point cloud data, and the formats may be JPEG, LAS, or PLY. For geological exploration data, point data (such as borehole data) can be converted from a CSV file to a Shapefile or GeoJSON through tools. For example, use the Geopandas library in Python to read the CSV file, specify the longitude and latitude fields, and export it as a Shapefile. Line data (such as geological profiles) and surface data (such as geological boundaries) can be converted from CAD files to Shapefiles or GeoJSON through QGIS or ArcGIS. The conversion of remote sensing data is mainly to convert JPEG or HDF formats to GeoTIFF. This process can be completed using the GDAL command-line tool or the "Raster->Conversion->Translate" tool in QGIS. For multi-band data (such as HDF), GDAL can be used to extract the required bands and convert them to GeoTIFF. The conversion of UAV aerial survey data includes image data and LiDAR point cloud data. Image data can be converted from JPEG to GeoTIFF using GDAL, while LiDAR point cloud data needs to use PDAL or CloudCompare to convert LAS files to raster formats (such as GeoTIFF). For example, use the PDAL command-line tool to convert point cloud data to raster through a configuration file.

[0127] Convert all data to a unified coordinate system (such as WGS84, EPSG:4326), and verify the alignment and integrity of the data in GIS software. Through the above steps, geological exploration data, remote sensing data, and UAV aerial survey data can be successfully converted to a unified format, providing a reliable basis for subsequent multi-source data analysis and modeling. Conversion unified basic formula:

[0128] x′ = a + bx + cy + dxy

[0129] y′ = e + fx + gy + hxy

[0130] Where: x and y are the coordinates in the original coordinate system. x′ and y′ are the coordinates in the transformed coordinate system. a, b, c, d, e, f, g, and h are transformation coefficients, which are obtained by calculating control points.

[0131] Geometric calibration of data with different resolutions is a key step to ensure the spatial alignment of multi-source data. By geometric correction, the spatial deviation caused by different sensors, platforms, or acquisition conditions is eliminated, enabling data with different resolutions to be accurately matched in the same coordinate system. Common geometric correction methods include polynomial correction, triangulated irregular network (TIN) correction, and least squares matching, etc. For geometric correction of data with different resolutions, the polynomial correction formula is as follows:

[0132] x′ = a0 + a1x + a2y + a3x 2 + a4xy + a5y 2

[0133] y′ = b0 + b1x + b2y + b3x 2 + b4xy + b5y 2

[0134] Where: x and y are the coordinates in the original image. x′ and y′ are the coordinates in the corrected image. a0, a1, a2, a3, a4, a5 and b0, b1, b2, b3, b4, b5 are polynomial coefficients, which are obtained by calculating control points.

[0135] The steps of geometric calibration are as follows: First, collect ground control points (GCPs), which can be ground landmark points with known coordinates or feature points extracted from high-precision reference images. The number and quality of control points directly affect the correction accuracy, and they usually need to be evenly distributed in the image. Second, establish a geometric correction model using control points. For example, in polynomial correction, use the coordinates of control points to calculate polynomial coefficients and map the pixel coordinates of the original image to the reference coordinate system. Finally, apply the correction model to resample the image to generate the corrected image. Resampling methods can choose the nearest neighbor method, bilinear interpolation, or cubic convolution interpolation, and the specific choice depends on the resolution of the data and application requirements.

[0136] Taking remote sensing images as an example, the steps of polynomial correction using GDAL tools are as follows: First, prepare a control point file (such as CSV format), which contains the original image coordinates and reference coordinates. Then, use the gdalwarp command of GDAL for correction.

[0137] Time series calibration is performed on multi-source data collected at different times, such as geological exploration data, remote sensing data, and UAV aerial photography data. The Fourier transform formula is as follows:

[0138]

[0139] Where: F(k) is the k-th frequency component in the frequency domain. f(n) is the n-th sampling point in the time domain. N is the total number of sampling points. k is the frequency index (k = 0, 1, 2, …, N - 1). I is the imaginary unit, satisfying i2 = -1. E is the natural constant, approximately equal to 2.71828. π is the pi, approximately equal to 3.14159.

[0140] Perform Fourier transforms on geological exploration data, remote sensing data, and UAV aerial photography data respectively, and extract their respective spectral features. By analyzing the amplitude and phase information of the spectra, identify the significant time periodic components in each data. Based on the spectral analysis results, calculate the time offset between different data. For example, if there is a phase difference between remote sensing data and UAV aerial photography data at a certain frequency component, the time offset can be calculated through the relationship between the phase difference and the frequency. For geological exploration data, if there is a time delay between its spectral features and other data, it can also be quantified through phase information. Using these time offsets, perform time correction on the data to ensure that the multi-source data is aligned on the time axis. For example, shift the time series of UAV aerial photography data forward or backward to make its time characteristics consistent with those of remote sensing data. After completing the time correction, verify the correction results. By recalculating the spectral features of the corrected data, check whether the phase difference between different data is significantly reduced or eliminated. In addition, combined with the actual application scenarios, such as geological activity monitoring or ecological environment assessment, verify the consistency and reliability of the corrected data. Finally, use the corrected multi-source data for comprehensive analysis to provide more accurate time series support for geological exploration, environmental monitoring, etc.

[0141] Step S102: Adopt an advanced data fusion algorithm to fuse the preprocessed multi-source data to form a comprehensive dataset of karst geological information.

[0142] Specifically, fuse the preprocessed multi-source data such as geological exploration data, remote sensing data, and UAV aerial photography. Weighted fusion is a simple and effective data fusion method. By assigning weights to different data sources and comprehensively considering the importance of each data source, a fusion result is generated. The specific steps are as follows:

[0143] 1. Data source feature extraction

[0144] Extract the eigenvalues from the preprocessed geological exploration data, remote sensing data, and UAV aerial photography data respectively. For example, the geological exploration data can extract features such as the depth and density of the underground structure, the remote sensing data can extract features such as the surface index and temperature, and the UAV aerial photography data can extract features such as the surface texture and elevation. Assume that the extracted eigenvalues are x1, x2, and x3 respectively.

[0145] 2. Weight Assignment

[0146] According to the accuracy, resolution of the data source or expert experience, assign weights w1, w2, and w3 to each data source, and satisfy w1 + w2 + w3 = 1. For example, the geological exploration data has high accuracy in underground structure information and can be given a larger weight (such as w1 = 0.5), the remote sensing data has an advantage in surface coverage information and can be given a medium weight (such as w2 = 0.3), and the UAV aerial photography data has high resolution in surface details and can be given a smaller weight (such as w3 = 0.2).

[0147] 3. Weighted Fusion Calculation

[0148] Use the weighted fusion formula to calculate the fused eigenvalue y:

[0149] y = w1·x1 + w2·x2 + w3·x3

[0150] For example, if x1 = 10 (eigenvalue of geological exploration data), x2 = 8 (eigenvalue of remote sensing data), and x3 = 6 (eigenvalue of UAV aerial photography data), then the fusion result is:

[0151] y = 0.5·10 + 0.3·8 + 0.2·6 = 5 + 2.4 + 1.2 = 8.6

[0152] 4. Result Generation

[0153] Perform weighted fusion on all eigenvalues to generate a comprehensive karst geological information dataset. This method is simple and intuitive and is applicable to scenarios where the characteristics of the data source are relatively clear and the weights are easy to determine.

[0154] Principal component analysis (PCA) is a commonly used dimensionality reduction method that can extract the main features from multi-source data, eliminate redundant information, and retain the key information of the data at the same time. The specific steps are as follows:

[0155] 1. Data Matrix Construction

[0156] Construct the preprocessed geological exploration data, remote sensing data, and UAV aerial photography data into a data matrix X, where each column represents the feature vector of a data source. Assume there are n samples and m features are extracted for each data source, then the dimension of the data matrix X is n×3m.

[0157] 2. Covariance Matrix Calculation

[0158] Calculate the covariance matrix C of the data matrix X:

[0159] C = (1 / n)XTX

[0160] The covariance matrix reflects the correlation between different features.

[0161] 3. Eigenvalue decomposition

[0162] Perform eigenvalue decomposition on the covariance matrix C to obtain eigenvalues and eigenvectors. The eigenvalues represent the variances of the principal components, and the eigenvectors represent the directions of the principal components.

[0163] 4. Principal component selection

[0164] According to the magnitudes of the eigenvalues, select the top k principal components (usually select the principal components with a cumulative variance contribution rate reaching a certain threshold, such as 95%). Construct the projection matrix P, whose column vectors are the top k eigenvectors.

[0165] 5. Data dimensionality reduction and fusion

[0166] Project the original data matrix X into a low-dimensional space to obtain the fused dataset Y after dimensionality reduction:

[0167] Y = X·P

[0168] For example, if the dimension of the original data matrix X is 100×9 (100 samples, 9 features), and the top 3 principal components are selected, then the dimension of the projection matrix P is 9×3, and the dimension of the dataset Y after dimensionality reduction is 100×3.

[0169] 6. Result generation

[0170] The dataset Y after dimensionality reduction is the fused comprehensive karst cave geological information dataset. PCA can effectively reduce the data dimension while retaining the main features of multi-source data, and is applicable to the fusion scenario of high-dimensional data.

[0171] Perform cross-validation on the fused dataset to evaluate its accuracy and reliability. Divide the dataset into a training set and a test set, use the training set to train the model, and verify it on the test set. For example, the K-fold cross-validation method can be adopted. Divide the dataset into K subsets, sequentially use each subset as the test set, and the remaining subsets as the training set, and repeat K times to calculate the average performance metrics of the model (such as accuracy, recall rate, F1 score, etc.). Through cross-validation, the stability and generalization ability of the fusion result can be comprehensively evaluated. For example, in the karst cave geological information dataset, if the performance metrics of the model fluctuate slightly on different subsets, it indicates that the fusion result has high reliability.

[0172] Compare the fused karst cave geological information with the on-site survey data to further verify its accuracy. For example, compare the characteristics such as the distribution, depth, and shape of karst caves in the fusion result with the known on-site survey data item by item, and calculate the consistency indices (such as mean square error, correlation coefficient, etc.). If the fusion result is highly consistent with the on-site data in key characteristics, it indicates that the fusion algorithm is effective; if there are significant differences, the reasons need to be analyzed, which may be due to unreasonable weight allocation of data sources or improper parameter settings of the fusion algorithm. For example, if the deviation of the karst cave depth in the fusion result from the on-site data is large, it may be necessary to adjust the weight of geological exploration data or optimize the fusion algorithm.

[0173] According to the verification results, adjust the parameters or weights of the fusion algorithm to further optimize the fusion effect. For example, in the weighted fusion method, if the weight of a certain data source is too high, resulting in a large deviation in the fusion result, the weights can be reallocated to make it more in line with the actual needs; in the PCA fusion method, if important features are lost after dimensionality reduction, the number of principal components can be increased or the feature extraction method can be adjusted. In addition, machine learning methods (such as grid search or Bayesian optimization) can be combined to automatically find the optimal parameter combination. Through multiple iterations of optimization, a high-quality comprehensive karst cave geological information dataset is finally formed, providing reliable data support for applications such as geological exploration and disaster warning. For example, the optimized dataset can more accurately predict the distribution of karst caves.

[0174] Step S103: Use machine learning algorithms or deep learning models to extract shape features, size features, location features, etc. of karst caves from the fused dataset.

[0175] Extract karst cave features based on a deep learning model. Deep learning models, especially convolutional neural networks (CNNs), have been widely used in the analysis of spatial data (such as remote sensing images, geological data) due to their powerful feature extraction capabilities. For the task of karst cave location extraction, CNNs can automatically learn spatial features from multi-source fused data and output accurate location features.

[0176] The input data is the fused dataset, including geological exploration data, remote sensing data, and UAV aerial photography data. These data can be represented as multi-dimensional matrices. For example: geological exploration data containing features such as depth and density, remote sensing data containing features such as terrain texture, and UAV aerial photography data containing features such as elevation and texture. Integrate multi-source data into a multi-dimensional tensor, for example, with a shape of n×h×w×c, where n is the number of samples, h and w are spatial dimensions (such as height and width), and c is the number of feature channels.

[0177] (1) Design a convolutional neural network (CNN) architecture for karst cave data analysis:

[0178] 1. At the input layer, receive the multi-dimensional tensor as input.

[0179] 2. In the convolutional layer, local features are extracted through the convolutional kernel. The formula for the convolution operation is:

[0180]

[0181] where wab is the convolutional kernel weight, b is the bias term, and σ is the activation function (such as ReLU).

[0182] 3. In the pooling layer, the dimension of the feature map is reduced through max pooling or average pooling. The formula for max pooling is: y ij = max(x i·s,j·s ,..., x i·s+k-1,j·s+k-1 )

[0183] where s is the stride and k is the pooling window size.

[0184] 4. In the fully connected layer, the features extracted by the convolutional layer and the pooling layer are mapped to the output space.

[0185] 5. In the output layer, the location features of the karst cave are output, such as longitude and latitude coordinates or relative positions.

[0186] The mean squared error (MSE) is used as the loss function to measure the difference between the predicted position and the true position:

[0187]

[0188] where yi is the true position and yi is the predicted position.

[0189] The Adam optimization algorithm is used to update the model parameters, and the learning rate decay strategy is combined to improve the training efficiency. Dropout or L2 regularization is used to prevent the model from overfitting.

[0190] The root mean squared error (RMSE) or mean absolute error (MAE) is used to evaluate the model performance:

[0191]

[0192] Specifically, a dataset of karst cave geological information in a certain area contains 1000 samples. Model training: 80% of the data is used as the training set and 20% of the data is used as the test set to train the CNN model. The RMSE of the model on the test set is 0.05 and the MAE is 0.03, indicating a high prediction accuracy.

[0193] (2) Feature Integration and Result Output

[0194] 1. Feature Integration Method

[0195] Feature integration is to uniformly represent the extracted cave shape features, size features, and location features to generate a complete key feature dataset of caves. The integration methods include: concatenating different feature vectors column by column to generate a comprehensive feature matrix. Assigning weights according to the importance of features to generate weighted feature vectors.

[0196] 2. Feature database construction

[0197] Design a relational database (such as MySQL) or a non-relational database (such as MongoDB) to store cave feature data. Create a table cave_features with fields shape, size, and location. Import the integrated feature data into the database for subsequent querying and analysis.

[0198] 3. Result visualization

[0199] Use tools such as Matplotlib, Seaborn, or Tableau for data visualization. Use a 3D graph to display the shape features, size features, and location of the caves. Use a heat map to display the distribution density of the caves. In the 3D graph, the X-axis and Y-axis represent the location of the caves, the Z-axis represents the depth of the caves, and the color represents the size of the caves.

[0200] Specifically, a dataset of cave geological information in a certain area contains 500 samples. Integrate the extracted cave shape features, size features, and location features into a comprehensive feature dataset. Import the integrated data into a MySQL database, and the table cave_features contains fields shape, size, and location.

[0201] Step S104: Construct a 3D model of the cave using the extracted cave features with a 3D modeling software or platform. At the same time, through a human-computer interaction method, finely adjust and optimize the model to ensure the accuracy and authenticity of the model. Step S104 specifically includes the following steps:

[0202] Step S501: Construct a high-precision cave model using 3D modeling technology through integrating, preprocessing, and feature extraction of multi-source data.

[0203] The extracted cave feature dataset includes shape features, size features, and location features. Shape features include the major axis, minor axis, and orientation angle of the cave. Size features include the diameter or volume of the cave. Location features include the longitude and latitude coordinates or relative position of the cave. Convert the feature data into a format supported by 3D modeling software, such as point cloud data (Point Cloud) or mesh data (Mesh Data).

[0204] Generate point cloud data based on the location and shape features of the cave. Assume the center coordinates of the cave are (xc , y c , z c ), with the major axis being a, the minor axis being b, and the direction angle being θ, the calculation formula for the point cloud coordinates (x, y, z) is as follows:

[0205] x = x c + a·cos(φ)·cos(θ) - b·sin(φ)·sin(θ)

[0206] y = y c + a·cos(φ)·cos(θ) + b·sin(φ)·sin(θ)

[0207] z = z c + c·sin(φ)

[0208] Among them, φ is the parameter angle (0 ≤ φ < 2π), and c is the height of the karst cave.

[0209] Convert the point cloud data into mesh data, and use a triangulation algorithm (such as Delaunay triangulation) to generate triangular patches. The formula for triangulation is:

[0210]

[0211] Among them, v1, v2, v3 are the vertex coordinates of the triangular patch.

[0212] Import the generated mesh data using 3D modeling software (such as Blender, AutoCAD, MeshLab) or platforms (such as Unity, UnrealEngine). Render the 3D model in the software, set the material and lighting effects to make the model more realistic.

[0213] Step S502: Through the visualization interface and interaction tools (such as 3D editing, parameter adjustment, and real-time feedback), combined with expert experience and user input, dynamically adjust and optimize the model.

[0214] 1. Interaction Interface Design

[0215] The interface provides basic operations such as translation, rotation, and scaling, as well as advanced functions such as vertex editing and patch adjustment. Interact with the model through the mouse, keyboard, or touch device.

[0216] 2. Model Adjustment

[0217] Vertex Editing: Select the vertices in the model and adjust their positions to optimize the model shape. Assume the original coordinates of vertex vi are (x i , y i , z i ), and the adjusted coordinates are (x i ′, y i′, z i ), then the vertex displacement vector is:

[0218] Δv i = (x i ′ - x i , y i ′ - y i , z i ′ - z i )

[0219] Patch adjustment: Select the triangular patches in the model and adjust their normal directions or areas. Assume the normal vector of the triangular patch is n and the adjusted normal vector is n′, then the normal rotation matrix is:

[0220]

[0221] where θ is the rotation angle.

[0222] 3. Real-time preview

[0223] During the adjustment process, preview the model changes in real time to ensure that the adjustment effect meets the expectations. Use the level of detail (LOD) technology to dynamically adjust the model details according to the viewing distance and improve the interaction fluency.

[0224] Step S503: Continuously optimize the model accuracy by iteratively adjusting the model parameters, comparing with the actual geological data, and using the cross-validation method, and ensure the high accuracy and reliability of the model through error analysis and reliability evaluation.

[0225] The parameters of the karst cave model include shape, size, volume, topological structure, physical properties, the number of vertices and polygons, position, etc.

[0226] 1. Model optimization

[0227] Use the Laplacian smoothing algorithm to smooth the model surface. Assume the set of neighbor vertices of vertex vi is N(i), then the coordinates of the smoothed vertex are:

[0228]

[0229] Use the edge collapse algorithm to simplify the model and reduce the number of triangular patches. Assume the edge e ij connects vertices v i and v j , and the coordinates of the new vertex after folding are:

[0230]

[0231] 2. Model verification

[0232] Check whether the geometric properties (such as volume, surface area) of the model are consistent with the original data. Assume the volume of the model is V and the surface area is S, then the calculation formulas are as follows:

[0233]

[0234] where T is the triangular facet. Visually compare the optimized model with the original data to check the accuracy and authenticity of the model.

[0235] 3. Result Output

[0236] Export the optimized model into common formats (such as OBJ, STL, FBX) for subsequent applications.

[0237] Step S105: Compare and verify the constructed karst cave model with the actual geological data, and optimize and adjust the model according to the verification results to improve the accuracy and reliability of the model. Step S105 specifically includes:

[0238] Step S601: Comprehensively evaluate the consistency between the model and the actual geological data by calculating indicators such as spatial error, morphological error, and volume error.

[0239] Define the indicators for comparing the model with the actual geological data, including the spatial error between the model and the actual geological data in terms of spatial position, the morphological error between the model and the actual geological data in terms of morphological features, and the volume error between the model and the actual geological data in terms of volume.

[0240] Use the Euclidean distance to calculate the spatial error between the model points and the actual geological data points:

[0241]

[0242] where (x m , y m , z m ) are the coordinates of the model points, and (x g , y g , z g ) are the coordinates of the actual geological data points.

[0243] Use the curvature or surface normal vector difference to calculate the morphological error:

[0244] E f = ||n m - n g ||

[0245] where n m is the surface normal vector of the model, and n g is the surface normal vector of the actual geological data.

[0246] Calculate the volume error between the model and the actual geological data:

[0247] E v = |V m - V g |

[0248] where V m is the model volume and V g is the actual geological data volume.

[0249] S602: Gradually improve the model accuracy by analyzing the error distribution characteristics and adjusting the model parameters using the iterative optimization method until the preset error threshold is met.

[0250] Plot the error distribution map and analyze the spatial distribution characteristics of the errors. Identify the areas with larger errors and determine the key points for optimization.

[0251] According to the error analysis results, adjust the model parameters. Optimize the model point positions using the least squares method:

[0252]

[0253] Adjust the model surface morphology using the curvature optimization algorithm; adjust the model scale according to the volume error.

[0254] Repeat the comparison, analysis, and adjustment process until the model accuracy meets the requirements. Set the iteration termination condition, for example:

[0255] max(E sp , E f , E v ) < ε

[0256] where ε is the preset error threshold.

[0257] S603: Through cross-validation and uncertainty analysis, comprehensively evaluate the generalization ability and reliability of the model to ensure that the optimized model has high accuracy and stability in practical applications.

[0258] Divide the geological data into a training set and a validation set, which are used for model optimization and validation respectively to evaluate the generalization ability of the model. Calculate the confidence interval or uncertainty range of the model to evaluate the reliability of the model. For example:

[0259]

[0260] where E i is the i-th error value and E is the average error.

[0261] S604: In the model output and application stage, output the optimized model in a standardized format (such as a grid model or point cloud data) and apply it to actual engineering or geological analysis, while collecting feedback information to further optimize the model performance.

[0262] In the model output stage, the optimized karst cave model is converted into a usable standardized format, such as a mesh model (Mesh) or point cloud data (Point Cloud), to ensure its wide compatibility and application in different software platforms and tools. The mesh model describes the surface morphology of the karst cave through triangular or quadrilateral patches, which is suitable for visualization, simulation, and analysis; the point cloud data records the spatial distribution of the karst cave in the form of discrete points, which is suitable for high-precision reconstruction and detail processing. By selecting the appropriate output format, ensure that the model can meet the specific requirements of subsequent engineering or geological analysis.

[0263] In the model application stage, the output karst cave model is deployed to actual engineering or geological analysis, such as karst cave stability assessment, groundwater flow simulation, tunnel design, or geological disaster prediction. By combining with the actual scenario, verify the performance of the model in actual application, and collect relevant data and feedback information. For example, in tunnel engineering, use the model to predict the impact of karst caves on construction; in geological analysis, study the formation mechanism and evolution law of karst caves through the model. The goal of this stage is to ensure that the model not only has high theoretical precision but also can play its value in actual application.

[0264] In the feedback and optimization stage, according to the feedback information in actual application, further adjust and optimize the model. For example, if there is a large deviation between the prediction result of the model in a certain area and the actual observation, it can be corrected by recalibrating parameters or introducing new data sources. Through continuous iterative optimization, improve the precision and reliability of the model, so that it can better serve actual engineering and geological research. This process reflects the dynamic nature and continuous improvement characteristics of model construction, and the ultimate goal is to achieve a high degree of fit between the model and actual needs.

[0265] The embodiment of the present invention also provides a rapid karst cave modeling system based on multi-source data, including:

[0266] A multi-source data acquisition and preprocessing module, which is used to collect multi-source data including geological exploration data, remote sensing data, and UAV aerial photography data, and preprocess the multi-source data;

[0267] A multi-source data fusion module, which is used to fuse the preprocessed data by using a weighted fusion or principal component analysis (PCA) method to form a comprehensive karst cave geological information data set:

[0268] A karst cave feature extraction module, which is used to extract shape features, size features, and position features from the fused data by using machine learning algorithms or deep learning models;

[0269] 3D modeling and interaction optimization module, which is used to convert the extracted karst cave features into point cloud data or mesh data, and construct a karst cave model using 3D modeling software; perform vertex editing and patch adjustment on the initial model through a human-computer interaction tool, and correct the model form in combination with expert experience;

[0270] Model verification and iterative optimization module, which is used to compare the constructed karst cave model with actual geological data, and calculate the spatial error, morphological error and volume error; if the error exceeds the preset threshold, iteratively adjust the model parameters until the accuracy requirement is met.

[0271] Through the organic combination of multi-source data fusion, intelligent analysis and 3D modeling technology, the present invention significantly improves the efficiency and accuracy of karst cave geological research: First, integrate multi-modal data such as geological exploration, remote sensing, and UAV aerial photography, break through the spatial and resolution limitations of a single data source, and comprehensively capture the complex features of the 3D morphology and geological environment of karst caves; Second, use machine learning / deep learning algorithms to realize the automatic extraction of karst cave features, replace the traditional manual analysis mode, reduce subjective bias and accelerate the processing flow; Third, through the full-process closed-loop mechanism of data calibration, human-computer collaborative optimization and iterative verification, ensure the high consistency between the 3D model and actual geological data, and effectively overcome the defect of poor generalization of traditional empirical models; Finally, based on UAV aerial photography and remote sensing technology, reduce the risks and costs of field operations, expand the application scenarios in combination with intelligent modeling tools, significantly improve the detection ability and model universality of complex karst cave systems, and provide a highly reliable and cost-effective technical solution for fields such as geological engineering and ecological protection.

[0272] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A rapid modeling method for karst caves based on multi-source data, characterized in that It includes the following steps: Step S101: Collect multi-source data including geological exploration data, remote sensing data, and UAV aerial photography data, and preprocess the multi-source data; Step S102: Fuse the preprocessed data using a weighted fusion or principal component analysis (PCA) method to form a comprehensive karst cave geological information dataset; Step S103: Use machine learning algorithms or deep learning models to extract shape features, size features, and location features from the fused data; Step S104: Convert the extracted karst cave features into point cloud data or mesh data, and use 3D modeling software to construct a karst cave model; perform vertex editing and patch adjustment on the initial model through a human-computer interaction tool, and correct the model form in combination with expert experience; Step S105: Compare the constructed karst cave model with the actual geological data, and calculate the spatial error, shape error, and volume error; if the error exceeds the preset threshold, iteratively adjust the model parameters until the accuracy requirement is met.

2. The method according to claim 1, wherein The geological exploration data includes borehole data, seismic data, and geological profiles, the remote sensing data includes satellite images and radar data, and the UAV aerial photography data includes LiDAR point clouds and high-resolution images.

3. The method according to claim 2, wherein The preprocessing of the multi-source data includes: Data cleaning: Remove noise through filtering methods, interpolate and fill in missing data, and correct incorrect data; Format unification: Format unification includes format conversion and coordinate system unification. Format conversion is to convert geological exploration data, remote sensing data, and UAV aerial photography data into a geospatial raster vector format; Data calibration: Based on ground control point pairs, spatially align the multi-source data to eliminate spatial deviations caused by sensors or platforms; align the time series of the multi-source data through spectral analysis to ensure the consistency of the data on the time axis.

4. The method according to claim 1, wherein The weighted fusion includes dynamically assigning weights according to the accuracy and resolution of the data sources to generate a comprehensive dataset; The PCA fusion includes extracting principal components through covariance matrix decomposition, eliminating redundant information, and constructing a dimensionality-reduced dataset.

5. The method according to claim 1, wherein Step S104 specifically includes: Step S501: Through integrating, preprocessing, and feature extracting the multi-source data, use 3D modeling technology to construct a high-precision karst cave model. The extracted karst cave feature dataset includes shape features, size features, and location features; the shape features include the major axis, minor axis, and orientation angle of the karst cave; the size features include the diameter or volume of the karst cave; the location features include the longitude and latitude coordinates or relative position of the karst cave; convert the feature data into point cloud data or mesh data supported by 3D modeling software; Among them, point cloud data is generated according to the location and shape characteristics of the karst cave. Assuming that the central coordinates of the karst cave are (x c , y c , z c ), the major axis is a, the minor axis is b, and the direction angle is θ. Then the calculation formula for the point cloud coordinates (x, y, z) is as follows: x = x c + a·cos(φ)·cos(θ) - b·sin(φ)·sin(θ) y = y c + a·cos(φ)·cos(θ) + b·sin(φ)·sin(θ) z = z c + c·sin(φ) where φ is the parameter angle, 0 ≤ φ < 2π, and c is the height of the karst cave; Convert the point cloud data into mesh data, and use a triangulation algorithm to generate triangular patches. The formula for triangulation is: where v1, v2, and v3 are the vertex coordinates of the triangular patch. Import the generated mesh data into 3D modeling software, render the 3D model in the software, and set the material and lighting effects; Step S502: Through the visualization interface and interaction tools, combined with expert experience and user input, dynamically adjust and optimize the model. The specific steps are as follows: (1) Interactive interface design The interface provides basic operations such as translation, rotation, and scaling, as well as advanced functions such as vertex editing and patch adjustment, and interacts with the model through a mouse, keyboard, or touch device; (2) Model adjustment Vertex editing: Select vertices in the model and adjust their positions to optimize the model shape. Assume the original coordinates of vertex vi are (x i , y i , z i ), and the adjusted coordinates are (x i ′, y i ′, z i ). Then the vertex displacement vector is: Δv i =(x i ′ - x i , y i ′ - y i , z i ′ - z i ) Patch adjustment: Select the triangular patches in the model and adjust their normal direction or area. Assuming the normal vector of the triangular patch is n and the adjusted normal vector is n′, the normal rotation matrix is: where θ is the rotation angle; (3) Real-time preview Preview the model changes in real time during the adjustment process to ensure that the adjustment effect meets expectations. Use the level of detail technology to dynamically adjust the model details according to the viewing distance; Step S503: Continuously optimize the model accuracy by iteratively adjusting the model parameters, comparing with the actual geological data, and using the cross-validation method, and ensure the high accuracy and reliability of the model through error analysis and reliability assessment. The cave model parameters include shape, size, volume, topological structure, physical properties, the number of vertices and polygons, and position. The specific steps are as follows: (1) Model optimization Smooth the model surface using the Laplace smoothing algorithm. Assume that the vertex v i has a neighbor vertex set N(i), then the coordinates of the smoothed vertex are: Simplify the model using the edge-collapse algorithm to reduce the number of triangular facets. Assume edge e ij connects vertices v i and v j , and the coordinates of the new vertex after folding are: (2) Model verification Check whether the geometric properties of the model are consistent with the original data. Assuming the volume of the model is V and the surface area is S, the calculation formula is: where T is the triangular patch. Visually compare the optimized model with the original data to check the accuracy and authenticity of the model; (3) Result output Export the optimized model into common formats, and the common formats include OBJ, STL, and FBX.

6. A rapid modeling system for karst caves based on multi-source data, characterized in that, Including: Multi-source data acquisition and preprocessing module, which is used to collect multi-source data including geological exploration data, remote sensing data, and UAV aerial survey data, and preprocess the multi-source data; Multi-source data fusion module, which is used to fuse the preprocessed data by using the weighted fusion or principal component analysis (PCA) method to form a comprehensive cave geological information dataset: Cave feature extraction module, which is used to extract shape features, size features, and position features from the fused data by using machine learning algorithms or deep learning models; 3D modeling and interaction optimization module, which is used to convert the extracted cave features into point cloud data or mesh data, and construct a cave model by using 3D modeling software; perform vertex editing and patch adjustment on the initial model through human-computer interaction tools, and correct the model shape in combination with expert experience; Model verification and iterative optimization module, which is used to compare the constructed cave model with the actual geological data, and calculate the spatial error, shape error, and volume error; if the error exceeds the preset threshold, iteratively adjust the model parameters until the accuracy requirement is met.

7. The system according to claim 6, wherein The geological exploration data includes borehole data, seismic data, and geological profiles. The remote sensing data includes satellite images and radar data. The UAV aerial survey data includes LiDAR point clouds and high-resolution images.

8. The system according to claim 7, wherein The multi-source data acquisition and preprocessing module preprocesses the multi-source data, specifically including: Data cleaning: Remove noise through filtering methods, interpolate and fill in missing data, and correct incorrect data; Format Unification: Format unification includes format conversion and coordinate system unification; Data Calibration: Based on ground control point pairs, multi-source data are spatially aligned to eliminate spatial deviations caused by sensors or platforms; through spectral analysis, the time series of multi-source data are aligned to ensure the consistency of data on the time axis; Format Conversion: Convert geological exploration data, remote sensing data, and LiDAR point cloud data into geospatial raster vector formats.

9. The system according to claim 6, wherein The multi-source data fusion module performs weighted fusion on the preprocessed data, including dynamically allocating weights according to the accuracy and resolution of data sources to generate a comprehensive data set; the multi-source data fusion module performs fusion on the preprocessed data using the principal component analysis (PCA) method, including extracting principal components through covariance matrix decomposition, eliminating redundant information, and constructing a reduced-dimensional data set.

10. The system according to claim 6, characterized in that, The three-dimensional modeling and interaction optimization module is specifically used for By integrating, preprocessing, and extracting features from multi-source data, a high-precision karst cave model is constructed using three-dimensional modeling technology, where the extracted karst cave feature data set includes shape features, size features, and position features; Shape features include the major axis, minor axis, and orientation angle of the karst cave; Size features include the diameter or volume of the karst cave; Position features include the longitude and latitude coordinates or relative positions of the karst cave; convert the feature data into point cloud data or mesh data supported by three-dimensional modeling software; Among them, point cloud data is generated according to the position and shape characteristics of the karst cave. Assuming that the central coordinates of the karst cave are (x c , y c , z c ), the major axis is a, the minor axis is b, and the direction angle is θ. Then the calculation formula for the point cloud coordinates (x, y, z) is as follows: x = x c + a·cos(φ)·cos(θ) - b·sin(φ)·sin(θ) y = y c + a·cos(φ)·cos(θ) + b·sin(φ)·sin(θ) z = z c + c·sin(φ) where φ is the parameter angle, 0 ≤ φ < 2π, and c is the height of the karst cave; Convert the point cloud data into mesh data and generate triangular patches using the triangulation algorithm. The formula for triangulation is: where v1, v2, and v3 are the vertex coordinates of the triangular patch. Import the generated mesh data into three-dimensional modeling software, render the three-dimensional model in the software, and set the material and lighting effects; Through the visualization interface and interaction tools, combined with expert experience and user input, the model is dynamically adjusted and optimized. The specific steps are as follows: (1) Interactive Interface Design The interface provides basic operations such as translation, rotation, and scaling, as well as advanced functions such as vertex editing and patch adjustment, and interacts with the model through a mouse, keyboard, or touch device; (2) Model Adjustment Vertex Editing: Select vertices in the model and adjust their positions to optimize the model shape. Assume the original coordinates of vertex vi are (x i , y i , z i ), and the adjusted coordinates are (x i ′, y i ′, z i ). Then the vertex displacement vector is: Δv i =(x' i -x i ,y' i -y i ,z' i -z i ) Patch Adjustment: Select the triangular patches in the model and adjust their normal directions or areas. Assume the normal vector of the triangular patch is n and the adjusted normal vector is n′, then the normal rotation matrix is: where θ is the rotation angle; (3) Real-time Preview Preview the model changes in real time during the adjustment process to ensure that the adjustment effect meets expectations. Use the level of detail technology to dynamically adjust the model details according to the viewing distance; By iteratively adjusting model parameters, comparing with actual geological data, and using the cross-validation method, continuously optimize the model accuracy, and ensure the high accuracy and reliability of the model through error analysis and reliability evaluation. The karst cave model parameters include shape, size, volume, topological structure, physical properties, the number of vertices and polygons, and position. The specific steps are as follows: (1) Model Optimization Smooth the model surface using the Laplace smoothing algorithm. Assume that the vertex v i has a neighbor vertex set N(i), then the smoothed vertex coordinates are: Simplify the model using the edge collapse algorithm to reduce the number of triangular faces. Assume edge e ij connects vertices v i and v j , and the coordinates of the new vertex after folding are: (2) Model Verification Check whether the geometric properties of the model are consistent with the original data. Assume the volume of the model is V and the surface area is S, then the calculation formula is: Among them, T is a triangular patch. The optimized model is visually compared with the original data to check the accuracy and authenticity of the model; (3) Result output Export the optimized model into common formats, and the common formats include OBJ, STL, and FBX.

Images