Geological BIM modeling and engineering quantity calculation method based on point cloud sparse optimization

Abstract

The invention provides a geological BIM modeling and engineering quantity calculation method based on point cloud sparse optimization, and relates to the technical field of geological engineering, and the method comprises the steps: obtaining high-resolution point cloud data through an unmanned aerial vehicle oblique photography technology, combining the high-resolution point cloud data with drilling exploration data, and carrying out the denoising, classification, segmentation, lithology discrimination and registration, thereby obtaining a point cloud point cloud model; a geological BIM model before excavation is constructed; area point cloud data are obtained again after excavation, the point cloud precision of the slope toe area is optimized through an inverse distance weighted interpolation algorithm, and a geological BIM model after excavation is generated; and in combination with a classified and optimized digital elevation model, detecting the elevation difference of the model before and after excavation pixel by pixel by adopting a robust elevation difference calculation method based on a neighborhood window, and accurately calculating the volume change of each lithologic region to obtain the engineering quantity. Through the point cloud sparse optimization and elevation change detection technology, the precision of the geological BIM model and the reliability of volume calculation are improved, and the method is widely suitable for geological engineering and earth volume calculation scenes.

Description

Technical Field

[0001] The present application relates to the field of geological engineering technology, and in particular to a geological BIM modeling and engineering quantity calculation method based on point cloud sparse optimization. Background Art

[0002] In geological engineering and excavation construction, accurate acquisition of the three-dimensional morphology, internal structure and excavation volume of geological bodies is the core foundation of construction planning and management. However, traditional geological information collection and modeling methods have significant deficiencies in accuracy, comprehensiveness and data integration capabilities. At the same time, the detection accuracy of local elevation changes is low, which further limits the construction of geological BIM models and the accuracy of engineering volume calculations, thus bringing many challenges to construction management and decision-making.

[0003] Existing geological BIM modeling and dynamic calculation methods for excavation engineering quantities mainly rely on point cloud data acquisition and traditional geometric modeling techniques. These methods are centered on point cloud data processing, digital elevation model (DEM) generation, and volume calculation, and are designed to model the three-dimensional morphology of geological bodies and perform statistical analysis of engineering quantities. Usually, multi-view point cloud data is collected through drone oblique photography technology or ground laser scanning technology, and a basic geological three-dimensional model is established in combination with drilling exploration data, and then mathematical calculations are used to complete the evaluation of excavation engineering quantities.

[0004] Although traditional methods have achieved geological BIM modeling and engineering quantity calculation to a certain extent, due to the poor adaptability of fixed algorithms to complex terrain, it is difficult to deal with the sparsity of point cloud data and the insufficient expression of key regional features such as the slope foot. In addition, the traditional fixed threshold elevation difference detection method has limited performance in nonlinear and non-stationary geological change scenarios, and often ignores the influence of internal structural characteristics of geological bodies and external environmental factors, resulting in limited prediction accuracy and engineering quantity calculation accuracy. Especially in the slope foot area and boundary feature areas during the excavation process, the insufficient accuracy of these models may cause significant engineering quantity statistical errors, further limiting the application value of geological BIM models in actual construction. Summary of the invention

[0005] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0006] To this end, the first purpose of this application is to propose a geological BIM modeling and engineering quantity calculation method based on point cloud sparse optimization.

[0007] The second purpose of this application is to propose a geological BIM modeling and engineering quantity calculation device based on point cloud sparse optimization.

[0008] To achieve the above-mentioned purpose, the first embodiment of the present application proposes a geological BIM modeling and engineering quantity calculation method based on point cloud sparse optimization, including:

[0009] Obtaining geological body image data and three-dimensional point cloud data using oblique photography technology, and associating and combining the three-dimensional point cloud data with drilling exploration data;

[0010] Preprocessing the three-dimensional point cloud data, and constructing a geological BIM model before excavation using the preprocessed three-dimensional point cloud data, image data and drilling exploration data, wherein the data preprocessing includes denoising, classification, segmentation, lithology differentiation and registration;

[0011] After the excavation work is completed, the oblique photography technology is used to obtain the point cloud data of the same area, and the geological model is constructed again. In view of the unclear and sparse point cloud recognition in the slope foot area, the inverse distance weighted interpolation algorithm is used to optimize the data accuracy and construct the post-excavation geological BIM model;

[0012] The geological BIM model before excavation is divided into regions according to the lithology, and the pixel-by-pixel elevation difference calculation method based on the robust elevation difference calculation method based on the neighborhood window is used to calculate the geological BIM model before and after excavation for the digital elevation models of the two stages before and after excavation.

[0013] Based on the elevation difference calculation results, the volume change of each lithology area is calculated to obtain the engineering volume of the entire excavation area.

[0014] Optionally, the method of acquiring geological body image data and three-dimensional point cloud data by using oblique photography technology, and associating and combining the three-dimensional point cloud data with drilling exploration data, includes:

[0015] Collect multi-view image data of geological bodies by using drone oblique photography technology, wherein the multi-view image data includes vertical images and oblique images;

[0016] Based on PIX4D software, the collected multi-view image data is imported and processed, dense point cloud data is generated through the software's automated workflow, the spatial geometric information of the geological body is extracted, and three-dimensional point cloud data containing high-precision three-dimensional coordinates is formed;

[0017] The drilling exploration data including drilling location, formation depth, lithology distribution and physical properties are collected and integrated to establish a geological body internal structure information database and associate it with the three-dimensional point cloud data.

[0018] Optionally, the three-dimensional point cloud data is preprocessed, and the pre-excavation geological BIM model is constructed using the preprocessed three-dimensional point cloud data, image data and drilling exploration data, including:

[0019] De-noising the three-dimensional point cloud data, using a statistical filtering algorithm to remove isolated points, noise points and irregular point groups to ensure the smoothness and accuracy of the point cloud data;

[0020] Based on a multi-layer dynamic graph convolutional neural network, the denoised point cloud data is classified, and geological features, vegetation, buildings and noise in the point cloud are extracted according to geometric features, providing a labeling and classification basis for the subsequent elimination of non-geological feature point clouds;

[0021] Based on the classified three-dimensional point cloud data, deep feature learning is used to segment the data, and feature areas related to the geological body are extracted, while point cloud data irrelevant to the geological body are eliminated;

[0022] By distinguishing the lithology of the pure geological point cloud data generated after segmentation, and combining the drilling exploration data and deep learning classification model, the point cloud data of the geological body is refined into different lithology areas;

[0023] The lithologically differentiated three-dimensional point cloud data is registered, and the three-dimensional point cloud data scanned multiple times is spatially aligned by an iterative closest point algorithm, and the coordinates are unified and integrated with the drilling exploration data to ensure that the point cloud and image data and geological drilling information are unified in the same coordinate system, eliminate spatial errors and integrate surface and underground information;

[0024] The pre-processed three-dimensional point cloud data, image data and drilling exploration data are used to construct a complete pre-excavation geological BIM model to express the three-dimensional shape, internal structure and attribute information of the geological body.

[0025] Optionally, the pre-processed three-dimensional point cloud data, image data and drilling exploration data are used to construct a complete geological BIM model before excavation, which is used to express the three-dimensional shape, internal structure and attribute information of the geological body, including:

[0026] Importing the pre-processed three-dimensional point cloud data and drilling exploration data into BIM modeling software;

[0027] The geometric features of the ground surface, including the slope surface, slope foot and fault area, are extracted through the three-dimensional point cloud data, and a digital elevation model DEM of the ground surface is generated; an underground layered model is established in combination with the drilling exploration data, and the stratum depth information in the borehole is connected with the point cloud surface model to form a unified three-dimensional geological BIM model of the ground surface and the underground;

[0028] Add necessary geological attributes to the generated BIM model to distinguish the excavation volume of different lithologies and strata. The geological attributes include density and volume parameters related to the lithology attributes of the strata and the excavation engineering requirements;

[0029] Output a complete geological BIM model before excavation.

[0030] Optionally, after the excavation work is completed, the point cloud data of the same area is obtained by using oblique photography technology, and the geological model is constructed again. In view of the unclear and sparse point cloud recognition in the slope foot area, the inverse distance weighted interpolation algorithm is used to optimize the data accuracy, and the post-excavation geological BIM model is constructed, including:

[0031] After the excavation work is completed, the drone tilt photography technology is used again to collect multi-view images of the excavated area, and the three-dimensional point cloud data of the geological body after the excavation is obtained by using the multi-view image collection data collected again;

[0032] Based on the 3D point cloud data of the excavated geological body, the inverse distance weighted interpolation algorithm is used to complete the elevation values ​​of the missing points in the sparse point cloud areas. In order to solve the problem of unclear and sparse point cloud recognition in the slope foot area, the inverse distance interpolation combined with the local trend fitting method is used to adjust the interpolation strategy and parameters.

[0033] The optimized 3D point cloud data was combined with the drilling exploration data before excavation, and the post-excavation geological BIM model was constructed using Revit modeling software.

[0034] Optionally, the three-dimensional point cloud data based on the excavated geological body is used to fill in the elevation values ​​of the missing points in the area with sparse point clouds using an inverse distance weighted interpolation algorithm, and to adjust the interpolation strategy and parameters by combining inverse distance interpolation with a local trend fitting method for unclear and sparse point cloud recognition in the slope foot area, including:

[0035] Density analysis is performed on the 3D point cloud data of the excavated geological body to identify sparse areas. For each point p i If the number of points in its neighborhood is less than the preset number of points, then p i The area is a sparse area;

[0036] For sparse areas, the inverse distance weighted interpolation algorithm is used to complete the elevation value Z(x,y) of the missing point. The expression is:

[0037]

[0038] In the formula, Z i is the neighborhood point p i The known elevation value, d i is the distance between the interpolation point and the neighborhood point, and p is the distance weight index;

[0039] From the point cloud data of the slope foot area obtained based on classification before excavation, mark the key areas where sparse or unclear features are expressed;

[0040] In the marked slope foot area, feature points in the sparse area neighborhood are selected, and the local trend equation is established using the linear regression fitting method;

[0041] According to the fitting trend result obtained after solving the local trend equation, the parameters of the inverse distance interpolation are optimized, and the preliminary elevation value of the trend fitting is combined with the interpolation result of the inverse distance weighted algorithm to generate the final interpolation elevation value according to the weight;

[0042] After the interpolation is completed, the neighborhood mean smoothing method is used to eliminate the abruptness of the interpolation area;

[0043] After completing the interpolation and optimization, the elevation values ​​of the interpolation area are compared with the original three-dimensional point cloud data of the geological body after excavation, and the accuracy of the interpolation area is verified by calculating the mean square error.

[0044] Optionally, the geological BIM model before excavation is divided into regions according to lithology, and the digital elevation models before and after excavation are subjected to pixel-by-pixel elevation difference calculation using a robust elevation difference calculation method based on a neighborhood window, including:

[0045] Marking different lithology areas according to the three-dimensional point cloud data before excavation that has been differentiated by lithology;

[0046] Based on the annotated lithological areas, the elevation data generated before and after excavation are used to calculate the elevation difference pixel by pixel using the neighborhood window method. For each pixel, the elevation data before and after are compared within the neighborhood window, and the minimum absolute value is selected as the true elevation difference. At the same time, the elevation difference threshold is set to mark the change points or unchanged points, and the pixel-level results are mapped to the lithological areas.

[0047] Optionally, based on the annotated lithological area, the elevation data generated before and after excavation are used to calculate the elevation difference pixel by pixel using a neighborhood window method. For each pixel, the elevation data before and after the excavation are compared within the neighborhood window, and the minimum absolute value is selected as the true elevation difference. At the same time, an elevation difference threshold is set to mark the change point or the unchanged point, and the pixel-level result is mapped to the lithological area, including

[0048] According to the characteristics of the geological body, the appropriate neighborhood window size is set, the center pixel of the window is selected, its coordinates are located in the elevation data of the post-excavation stage, and all pixel values ​​corresponding to the window range in the elevation data of the pre-excavation stage are extracted;

[0049] For each pixel in the window, calculate the elevation difference between the central pixel after excavation and the neighboring pixels before excavation, and take the minimum absolute value as the true elevation difference of the current pixel;

[0050] Each pixel is marked as changed by setting an elevation difference threshold. If the actual elevation difference of the current pixel is greater than or equal to the elevation difference threshold, it is marked as a changed point, otherwise it is marked as an unchanged point.

[0051] The proportion of change points in each lithology area is counted, and the proportion of change pixels to total pixels is calculated. If the proportion of change points in a lithology area reaches a preset threshold, the area is marked as a change area, otherwise the area is marked as an unchanged area.

[0052] Optionally, the volume change of each lithology area is calculated based on the elevation difference calculation result to obtain the engineering quantity of the entire excavation area, including:

[0053] Based on the elevation difference calculation results, each lithological area is divided into regular grid cells;

[0054] For each grid cell, the volume change of a single grid is calculated based on the elevation difference and the grid area;

[0055] The volume changes of all grids in each lithology area are accumulated to obtain the volume change of the area. For the negative elevation difference area, the volume is counted as the excavation volume; for the positive elevation difference area, the volume is counted as the accumulation volume.

[0056] The volume change data of all lithology areas are summarized and the total engineering volume of the entire excavation area is calculated.

[0057] To achieve the above-mentioned purpose, the second embodiment of the present application proposes a geological BIM modeling and engineering quantity calculation device based on point cloud sparse optimization, comprising:

[0058] A data acquisition module is used to acquire geological body image data and three-dimensional point cloud data using oblique photography technology, and to associate and combine the three-dimensional point cloud data with drilling exploration data;

[0059] A pre-excavation geological model construction module is used to perform data preprocessing on the three-dimensional point cloud data, and to construct a pre-excavation geological BIM model using the preprocessed three-dimensional point cloud data, image data and drilling exploration data. The data preprocessing includes denoising, classification, segmentation, lithology differentiation and registration.

[0060] The post-excavation geological model reconstruction module is used to obtain point cloud data of the same area using oblique photography technology after the excavation work is completed, and to re-construct the geological model. In view of the unclear and sparse point cloud recognition in the slope foot area, the inverse distance weighted interpolation algorithm is used to optimize the data accuracy and construct the post-excavation geological BIM model.

[0061] The elevation difference calculation module is used to divide the geological BIM model before excavation into regions according to the lithology, and to calculate the pixel-by-pixel elevation difference of the geological BIM model before and after excavation using a robust elevation difference calculation method based on a neighborhood window for the digital elevation models of the two stages before and after excavation.

[0062] The engineering quantity calculation module is used to calculate the volume change of each lithology area based on the elevation difference calculation results to obtain the engineering quantity of the entire excavation area.

[0063] The technical solution provided by the embodiments of the present application brings at least the following beneficial effects:

[0064] By introducing point cloud sparse optimization technology, combined with the inverse distance weighted interpolation (IDW) algorithm and the local trend fitting method, the problems of sparsity of point cloud data and insufficient expression of slope foot area features in traditional methods are effectively solved, greatly improving the accuracy and detail expression capabilities of the geological model.

[0065] By adopting a robust elevation difference calculation method based on a neighborhood window and combining it with a classification-optimized digital elevation model (DEM), precise detection of elevation changes on a pixel-by-pixel basis is achieved, significantly improving the accuracy of excavation volume calculation, especially in complex terrain and non-stationary geological change scenarios.

[0066] By improving the data preprocessing process, including denoising, classification, segmentation and lithology differentiation, the impact of point cloud data registration errors on engineering quantity calculations can be effectively reduced, thereby improving the reliability and stability of the results.

[0067] By combining drone oblique photography technology to obtain high-resolution point cloud data, data collection becomes more efficient and convenient. Compared with traditional methods, this solution provides more comprehensive and accurate geological information support for construction planning and management, significantly improving the overall efficiency and decision-making ability of the project.

[0068] This technical solution has been specially optimized for the slope foot area and boundary feature areas under complex geological conditions, significantly reducing the statistical errors of traditional methods in these key areas and ensuring the practical application value of the geological BIM model and excavation quantity calculation results.

[0069] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0071] Figure 1A schematic diagram of a flow chart of a geological BIM modeling and engineering quantity calculation method based on point cloud sparse optimization provided in an embodiment of the present application;

[0072] Figure 2 A structural schematic diagram of a geological BIM modeling and engineering quantity calculation device based on point cloud sparse optimization provided in an embodiment of the present application. DETAILED DESCRIPTION

[0073] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0074] In view of the problems existing in the prior art, the present application provides a geological BIM modeling and engineering quantity calculation method based on point cloud sparse optimization. Figure 1 The present invention provides a flow chart of a geological BIM modeling and engineering quantity calculation method based on point cloud sparse optimization. Figure 1 As shown, the method comprises the following steps:

[0075] S1. Obtain geological image data and three-dimensional point cloud data using oblique photography technology, and associate the three-dimensional point cloud data with drilling exploration data.

[0076] In the embodiment of the present application, the high-precision three-dimensional point cloud and image data of the geological body are obtained by using oblique photography technology, and the drilling exploration data (including strata, lithology, depth, etc.) are combined to provide complete multi-source input data for the construction of the geological BIM model. The combined application of oblique photography and drilling exploration can ensure the comprehensiveness and accuracy of the data, laying a solid foundation for subsequent model generation.

[0077] Specifically, step S1 includes the following steps:

[0078] S101. Collect multi-view image data of geological bodies by using drone oblique photography technology, where the multi-view image data includes vertical images and oblique images.

[0079] In the embodiment of the present application, a drone equipped with a high-resolution camera is used to perform multi-angle oblique photography of the target geological body. During the image acquisition process, a combination of vertical shooting and oblique shooting is adopted to ensure that the acquired image data covers the entire range of the geological body, including key areas such as the slope surface, slope foot, and boundary features.

[0080] It is understandable that the multi-view acquisition method can effectively avoid the problem that traditional single-view image data is difficult to cover complex terrain, and ensure the integrity and accuracy of the generated 3D point cloud data. At the same time, in order to reduce the impact of the external environment (such as weather and lighting conditions) on the quality of data acquisition, it is recommended to choose clear and windless weather for aerial photography, and pre-plan the drone's route and shooting interval to ensure data uniformity and coverage. After the acquisition is completed, an image data set containing multiple angles and multiple resolutions is generated, providing a reliable data foundation for subsequent 3D point cloud generation.

[0081] S102. Based on PIX4D software, the collected multi-view image data is imported and processed, and dense point cloud data is generated through the software's automated workflow to extract the spatial geometric information of the geological body and form three-dimensional point cloud data containing high-precision three-dimensional coordinates.

[0082] In the embodiment of the present application, the collected multi-view image data is imported into the PIX4D software, and the built-in automated workflow is used to complete the image processing and point cloud data generation. The specific operation includes the following key steps:

[0083] First, the software uses image feature matching algorithms (such as the SIFT algorithm) to fully automatically align and stitch multi-view image data to construct a spatial geometric framework model of the geological body.

[0084] Secondly, after the preliminary registration is completed, dense point cloud data is generated based on the multi-view stereo reconstruction algorithm (MVS). The point cloud data contains the three-dimensional coordinate information of the geological body and the surface detail features, with high resolution and high precision.

[0085] Finally, to further improve the quality of point cloud data, the built-in filtering tool of PIX4D software is used to remove abnormal points and noise points and retain the real point cloud information of the geological surface.

[0086] Through the above processing steps, a dense point cloud dataset containing high-precision three-dimensional coordinates is finally generated. This data can fully reflect the external geometric characteristics of the geological body and lay the foundation for the subsequent construction of the geological BIM model.

[0087] S103. Collect and integrate drilling exploration data, establish a geological body internal structure information database, and associate it with the three-dimensional point cloud data. The drilling exploration data includes drilling location, formation depth, lithology distribution and physical properties.

[0088] In the embodiment of the present application, the drilling exploration data is an important data source reflecting the internal structural characteristics of the geological body, and its collection and integration includes the following contents:

[0089] First, obtain the distribution location information of the drill holes in the target area and collect relevant attribute data, including physical characteristics such as formation depth, lithology distribution, porosity, density, permeability, etc. All collected data should be derived from reliable exploration work records to ensure their accuracy and availability.

[0090] Then, based on the collected drilling data, a digital information database of the internal structure of the geological body is established, and the data is managed in layers. For example, data can be stored by different strata or lithology categories to facilitate subsequent calls and analysis.

[0091] Finally, the borehole exploration data is associated and integrated with the 3D point cloud data generated by oblique photography. Specifically, the geographic location of the borehole is aligned with the 3D coordinates of the point cloud data through the coordinate matching method to form a complete expression model of the external geometric features and internal structural characteristics of the geological body. This step can achieve the organic integration of multi-source data and provide comprehensive information support for the geological BIM model.

[0092] In addition, the integrated data needs to be preliminarily sorted out to remove overlapping images and abnormal points in the point cloud data, and to check the accuracy of the drilling exploration data.

[0093] S2. Preprocess the 3D point cloud data and use the preprocessed 3D point cloud data, image data and drilling exploration data to build a geological BIM model before excavation. Data preprocessing includes denoising, classification, segmentation, lithology differentiation and registration.

[0094] In order to ensure the integrity and accuracy of the 3D point cloud data, the data needs to be fully preprocessed before it is used to build the geological BIM model.

[0095] The core goal of data preprocessing is to clean, optimize and structure point cloud data, and integrate it with image data and drilling exploration data. The preprocessing steps include denoising, classification, segmentation, lithology differentiation and registration to ensure the consistency and relevance of point cloud data in terms of spatial geometric information and internal lithology characteristics.

[0096] Preprocessing can effectively improve data quality, eliminate noise and redundant points, and accurately express the geometric shape and structural characteristics of geological bodies, thus providing a solid foundation for the subsequent construction of geological BIM models. Finally, the construction of the geological BIM model before excavation is completed using preprocessed 3D point cloud data, image data, and drilling exploration data, providing an accurate basic model for geological analysis and engineering planning.

[0097] Specifically, step S2 also includes:

[0098] S201, denoising the three-dimensional point cloud data, using a statistical filtering algorithm to remove isolated points, noise points and irregular point groups, to ensure the smoothness and accuracy of the point cloud data.

[0099] First, the original point cloud data is identified and cleaned of noise points. Since the point cloud data generated by drone oblique photography may contain redundant environmental points (such as trees, equipment, etc.) or abnormal noise points, it is necessary to remove isolated points and error points through a filtering algorithm based on distance, density and normal vector distribution. This process can significantly improve the authenticity and usability of point cloud data and provide clean basic data for subsequent processing.

[0100] In the embodiment of the present application, a statistical outlier removal (SOR) algorithm is used to perform denoising on the original point cloud data to remove isolated points, noise points and irregular point groups.

[0101] Statistical filtering analyzes the neighborhood distribution of each point in the point cloud, calculates the distance from each point to the center of its neighborhood, and compares the distance with global statistical values ​​(such as mean and standard deviation) to identify and remove abnormal points that deviate from the normal range. At the same time, the main geometric feature points are retained to ensure data integrity. De-noising can significantly improve the smoothness and accuracy of point cloud data and effectively eliminate the interference of external environment (such as noise or equipment error) on data quality. The processed point cloud data has a higher signal-to-noise ratio, laying a high-quality data foundation for subsequent classification, segmentation, and registration operations.

[0102] Furthermore, the specific operation of step S201 includes the following detailed process:

[0103] S2011, firstly, the point cloud data is represented as a point set P = {p i ∣p i =(x i ,y i ,z i )}, where each point p i Contains its three-dimensional coordinate information (x, y, z). In order to identify isolated points and noise points, the embodiment of the present application performs i Contains the definition of a local neighborhood N i , the neighborhood range is determined by the set search radius r:

[0104] N i ={p j ∣∥p i -p j ∥≤r,p j ∈P}

[0105] Among them, ∥p i -p j∥ represents point p i and p j The Euclidean distance between the two points, r is the radius parameter used for neighborhood search, and the neighborhood size can be controlled by adjusting r. The purpose of local neighborhood partitioning is to associate each point with a set of points around it for subsequent statistical analysis.

[0106] S2012, for each point p i , calculate the average distance of its neighborhood points

[0107]

[0108] Among them, |N i | is point p i The total number of points in the neighborhood. That is, the total number of points in the neighborhood. By calculating the average distance from each point to the points in its neighborhood, the spatial distribution characteristics of the point in its local environment can be obtained.

[0109] S2013. Perform statistical analysis on the average distance of all points and calculate the global mean μ and standard deviation σ:

[0110]

[0111]

[0112] Where n is the total number of points in the point cloud data. The global mean μ reflects the average neighborhood distance of all points, while the standard deviation σ indicates the degree of dispersion of the distance distribution between points.

[0113] S2014. Based on the calculated global mean μ and standard deviation σ, set filtering conditions to remove isolated points and noise points. The filtering conditions are:

[0114]

[0115] Here, k is usually set to 2 to define an interval range. i The average neighborhood distance If it exceeds the range, the point is considered to be an isolated point or a noise point and will be removed.

[0116] After the above steps, the abnormal points in the point cloud data are removed, and the retained point cloud data is smoother and more accurate, providing high-quality basic data for subsequent classification, segmentation and modeling operations.

[0117] S202. Based on a multi-layer dynamic graph convolutional neural network, the denoised point cloud data is classified, and geological features, vegetation, buildings and noise in the point cloud are extracted according to geometric features, providing a labeling and classification basis for the subsequent elimination of non-geological feature point clouds.

[0118] In the embodiment of the present application, a multi-layer dynamic graph convolutional neural network (DGCNN) is used to classify point cloud data. The multi-layer dynamic graph convolutional neural network can capture the local and global features of the point cloud through the dynamic neighborhood feature map and adapt to complex geometric forms. The classification process extracts key information from the geometric features of the point cloud and divides the point cloud data into categories such as geological feature points, vegetation points, building points, and noise points.

[0119] Specifically, the multi-layer dynamic graph convolutional neural network first constructs a neighborhood feature map, extracts local geometric relationships through convolution operations, and then captures global spatial information through multi-layer deep network superposition. This method can dynamically adjust the classification criteria to ensure the robustness and accuracy of the classification results. The classified data provides a clear basis for the subsequent elimination of non-geological feature point clouds and the extraction of key geological areas.

[0120] Furthermore, the specific operation of step S202 includes the following detailed process:

[0121] S2021, using the denoised point cloud data obtained in step S201 as input, and using the edge convolution function to extract features of the geometric relationship between each point in the point cloud and its neighboring points. First, define the edge feature e between the point and its neighboring points ij , and its calculation formula is:

[0122] e ij =ReLU(W(p j -p i )+b)

[0123] Among them, p i is the current point, p j is its neighborhood point, W and b are learnable parameters, and ReLU is the activation function used to extract nonlinear local features. The edge feature describes the point p i With neighboring point p j The relative relationship between them can capture the local geometric characteristics of the point cloud.

[0124] S2022. For each point p i , aggregate the edge features in its neighborhood through the maximum pooling operation and extract local features

[0125]

[0126] Among them, l represents the number of layers of edge convolution. Through this process, point p iIt is possible to obtain high-dimensional local geometric feature information from its neighborhood points. This operation is repeated many times, and by stacking multiple layers of convolutional networks, local features of different scales of the point cloud are extracted layer by layer to ensure that the details and global geometric information of the point cloud are captured.

[0127] S2023. Through the multi-layer dynamic graph convolutional network (DGCNN), local features are extracted layer by layer as deep geometric features. The l+1th layer feature is represented as:

[0128] H (l+1) =f(H (l) )

[0129] Among them, H (l) is the feature of the lth layer, and f is the ReLU activation function. The multi-layer DGCNN structure can dynamically adjust the edge features between points and neighboring points to extract richer geometric information.

[0130] S2024. Based on the extraction of local features, the global maximum pooling operation is used to perform global feature aggregation on the deep features of the point cloud data to generate a global feature vector h global :

[0131]

[0132] Among them, H(L) is the feature matrix of the last layer of the network, and the max operation takes the maximum value of the feature of each point to ensure the disordered characteristics of the point cloud. The global feature vector h global It can represent the overall geometric characteristics of point cloud data and contains global geometric information.

[0133] S2025. The global feature vector h global Input the fully connected layer and use the softmax function to output the classification probability distribution

[0134]

[0135] Among them, the softmax function is used to calculate the probability distribution of each point belonging to different categories (such as geological bodies, vegetation, buildings, noise, etc.), and W and b are the weights and bias parameters of the fully connected layer. The classification result can assign a category label to each point in the point cloud data, providing a clear basis for subsequent point cloud segmentation.

[0136] Through the above steps, DGCNN can effectively capture the local and global features of point cloud data, achieve high-precision classification of geological bodies, vegetation, buildings, noise and other categories in the point cloud, and provide important data support for the further processing of geological models.

[0137] S203, based on the classified three-dimensional point cloud data, segment it using deep feature learning to extract feature areas related to the geological body, and at the same time eliminate point cloud data that is not related to the geological body.

[0138] In an embodiment of the present application, a deep feature learning algorithm is used to segment geologically related areas based on the classified point cloud data.

[0139] During the segmentation process, the initial seed points related to the geological features are first selected, and then expanded to the entire target area through the region growing algorithm to extract key geological areas such as slopes, slope toes, and fault planes. At the same time, the high-dimensional features extracted by the deep learning model are used to further optimize the regional boundaries to ensure the integrity and coherence of the segmentation results. For irrelevant point cloud data (such as trees, buildings, etc.), they are removed during the segmentation process to improve the purity of geological body-related data. The segmented data can better express the three-dimensional structural characteristics of the geological body and reduce the interference of redundant points on subsequent modeling and analysis, thereby providing high-quality data support for the construction of geological BIM models.

[0140] Furthermore, the specific operation of step S203 includes the following detailed process:

[0141] S2031. Use the deep features H of the point cloud data to construct a feature relationship matrix R, which is used to describe the similarity relationship between each point in the point cloud. Feature relationship matrix R ij The calculation is based on the cosine similarity function:

[0142]

[0143] Among them, R ij Represents point p i and point p j The feature similarity, h i and h j They are point p i and point p j The deep features of , ∥·∥ represents the L2 norm of the feature vector. Cosine similarity can effectively measure the similarity of features in point clouds and provide support for subsequent classification and segmentation.

[0144] S2032, according to the similarity matrix R and the classification result, each point p in the point cloud data i Assign classification labels The classification label assignment formula is as follows:

[0145]

[0146] Among them, argmax means selecting the classification label category with the greatest similarity. The classification labels include geological feature areas (such as slope, slope foot, fault surface, etc.) and non-geological feature areas (such as trees, houses, noise points). Through this process, each point in the point cloud is clearly marked as the corresponding category.

[0147] S2033. Extract the point cloud related to the geological body as a feature region according to the assigned category label. Specifically, extract the slope features according to the consistency of the local normal vector and the curvature information, extract the slope foot features using the boundary points of the region with large geometric feature changes, and extract the fault feature points according to the region with high local feature gradient. The formula for extracting the feature region is:

[0148]

[0149] in, It is a local gradient used to identify different characteristic areas. The combination of curvature information and gradient changes ensures the accuracy of the extraction results and can effectively distinguish geological characteristic areas such as slope surface, slope foot and fault.

[0150] S2034. Based on the classification results, remove points marked as non-geological objects (such as vegetation, buildings, and noise points) in the point cloud to generate a pure point cloud dataset. geo is defined as follows:

[0151]

[0152] The final point cloud dataset only retains the point clouds related to the geological body, and removes irrelevant points such as trees, houses and other noise points. Through this segmentation process, a high-purity geological feature point cloud is generated, providing high-quality data support for subsequent geological BIM model construction and geological analysis.

[0153] S204, by distinguishing the lithology of the pure geological point cloud data generated after segmentation, combining the drilling exploration data and the deep learning classification model, the point cloud data of the geological body is refined into different lithology areas.

[0154] In the embodiment of the present application, the segmented geological point cloud data is differentiated by lithology by combining the drilling exploration data and the deep learning classification model. The geological body point cloud is refined into multiple lithology areas by matching the lithology distribution characteristics (such as sandstone, shale, clay, etc.) in the drilling exploration data with the geometric characteristics of the point cloud. The lithology differentiation gives the geological BIM model physical and chemical property information, so that the model not only has three-dimensional morphological characteristics, but also can reflect the diversity and distribution of lithology.

[0155] Furthermore, the specific operation of step S204 includes the following detailed process:

[0156] S2041, extracting the clean point cloud data P generated in step S203 geo , obtain lithology distribution information from drilling exploration data. The drilling data includes the three-dimensional coordinates (x, y, z) of the drilling location and the corresponding lithology category label y rock (such as soil layer, sandstone, shale, etc.). For each point p in the point cloud i , calculate the distance d to the drilling position, assign preliminary lithology labels to the point cloud points, and the classification rules are:

[0157]

[0158] Among them, r is the borehole influence radius, which is used to define the effective range of the borehole on the point cloud points. Unclassified points represent point clouds that are far away from all boreholes and will be further processed in subsequent steps. This step gives the point cloud data preliminary lithology labels based on spatial distribution.

[0159] S2042, using the feature relationship matrix R of step S2031 to reconstruct the point cloud feature similarity matrix. Using this matrix as input, a multi-layer dynamic graph convolutional network (DGCNN) is used to extract the local and global features required for lithology classification. The features of each point cloud point are dynamically extracted through the edge convolution layer, and combined with the geometric and attribute information of the surrounding neighborhood to generate deep features for classification. For each point, DGCNN uses the classification probability distribution P i Predict the probability of belonging to each lithology category. Finally, assign the final lithology category to the point cloud point according to the maximum value of the probability distribution:

[0160]

[0161] Among them, P i For point p i The probability distribution of different lithology categories, argmax is used to select the category with the highest probability value as the lithology label of the point cloud point. DGCNN ensures the accuracy of classification results by dynamically adjusting neighborhood features, especially in complex geological environments.

[0162] S2043, mapping the lithology classification results to the point cloud data points, and outputting the three-dimensional volume and boundary information of each lithology area. Specifically, the spatial distribution range of each lithology category is calculated using the classified point cloud, and the boundary points are extracted to define the regional boundaries, and the volume information of each lithology area is calculated based on the classification results.

[0163] Ultimately, the output lithology classification results and regional volumes provide accurate data support for subsequent engineering quantity calculations, and also provide an important basis for the attribute assignment of the geological BIM model.

[0164] S205. Register the lithologically differentiated 3D point cloud data, spatially align the 3D point cloud data scanned multiple times through an iterative nearest point algorithm, and unify and fuse the coordinates with the drilling exploration data to ensure that the point cloud and image data as well as geological drilling information are unified in the same coordinate system, eliminate spatial errors, and integrate surface and underground information.

[0165] In the embodiment of the present application, the iterative closest point algorithm is used to accurately register the point cloud data and the drilling exploration data of multiple scans. The registration process first uses a global feature matching method (such as coarse registration based on feature points) to perform preliminary alignment on the multiple scanned point clouds. Then, the iterative closest point algorithm is used to gradually minimize the distance error between the point clouds, and high-precision registration is achieved through multiple rounds of iterations.

[0166] In addition, in order to ensure the consistency of multi-source data, the embodiment of the present application also unifies the coordinate system and spatially integrates the point cloud data with the drilling exploration data, integrating the surface geometric features with the underground lithology information. The registration operation can effectively eliminate the spatial errors between multi-source data, making the three-dimensional information of the geological body more complete, and providing an accurate data foundation for the subsequent construction of the geological BIM model.

[0167] Furthermore, the specific operation of step S205 includes the following detailed process:

[0168] S2051, select a set of control points P with significant geological features from the multiple scan point cloud data control , and set the reference point cloud P ref and the target point cloud P target , complete the preliminary alignment by rigid transformation of the control points, and calculate the rough transformation matrix T init The transformation matrix solution process is achieved by minimizing the distance error between the reference point cloud and the target point cloud control point. The formula is:

[0169]

[0170] Among them, t(p i ) represents the transformation matrix t of the point p in the reference point cloud i The position after the initial rigid transformation, q i is the position of the corresponding point in the target point cloud. This step completes the rough alignment of the point cloud and provides a basis for subsequent precise registration.

[0171] S2052: Based on the result of the rough alignment, the registration accuracy of the point cloud is further optimized using the iterative closest point (ICP) algorithm. The ICP algorithm iteratively optimizes the transformation matrix T by repeatedly matching the nearest neighbor point pairs in the point cloud. opt, and finally achieve accurate registration of the two sets of point clouds. The formula is:

[0172]

[0173] Among them, T(p i ) is the target point after transformation, q i is the reference point cloud with p i The nearest neighbor point. Through multiple rounds of iterations, the error between point clouds is gradually reduced to ensure high-precision registration of point clouds.

[0174] S2053, map the image data generated by oblique photography to the point cloud surface, and use the three-dimensional coordinates (x, y, z) of the point cloud and the projection matrix M of the image to calculate the two-dimensional coordinates (u, v) of each point in the image:

[0175]

[0176] Among them, M is the projection matrix of the image, and (u, v) is the corresponding position of the point cloud point on the image. Through this process, the texture information of the image is mapped to the point cloud surface, the image texture of the point cloud surface is enhanced, and a textured point cloud is formed to provide support for the visualization of the geological BIM model.

[0177] S2054, using the drilling position data as control points, aligning the point cloud data with the drilling exploration data to unify them into the same coordinate system. By minimizing the error between the point cloud control points and the drilling control points, the rigid transformation matrix E is calculated. align :

[0178]

[0179] Among them, T(p j ) is the position of the transformed point cloud control point, p j ′ is the corresponding control point position in the drilling data. Through error optimization, the coordinate consistency of the point cloud data and the drilling data can be ensured, further improving the registration accuracy of multi-source data.

[0180] S2055. Integrate the registered point cloud data, image texture and drilling data into a unified three-dimensional data set.

[0181] The final integrated data set contains surface geometric features, image texture information, and borehole lithology information, which can form a complete three-dimensional geological information expression. The integrated three-dimensional data set provides comprehensive data support for subsequent geological BIM model construction and analysis.

[0182] S206. Using the pre-processed 3D point cloud data, image data and drilling exploration data, a complete pre-excavation geological BIM model is constructed to express the 3D shape, internal structure and attribute information of the geological body.

[0183] After completing the preprocessing of the point cloud data, the processed point cloud data, image data and drilling exploration data are combined to build a complete geological BIM model. During the construction process, the point cloud data is first used to generate the three-dimensional geometric shape of the geological body to ensure the high-precision expression of the surface features of the model; then, combined with the results of lithology differentiation, the lithology distribution and related physical properties (such as density, permeability, etc.) are given to the model to enrich its internal attribute information; finally, by integrating the surface point cloud and underground drilling data, a geological BIM model with complete three-dimensional spatial expression is formed. The final geological BIM model can not only display the external shape of the geological body, but also realize the accurate expression of the geological body in combination with the lithology distribution and internal structure, providing comprehensive data support for subsequent engineering planning, construction management and dynamic monitoring.

[0184] Specifically, the specific operation of step S206 includes the following detailed process:

[0185] S2061. Import the pre-processed point cloud data and drilling exploration data into BIM modeling software (such as Revit), which is not specifically limited in this application.

[0186] Point cloud data is mainly used to construct a three-dimensional surface model, including geometric information of characteristic areas such as slopes and slope toes, while borehole exploration data provides layered information of underground strata, especially the depth and spatial distribution of lithology. The combination of point cloud and borehole exploration data can ensure that the geological model covers complete information on both the surface and underground, providing comprehensive data support for further construction of the model.

[0187] S2062. Extract the geometric features of the ground surface, including the slope surface, slope foot and fault area, through the point cloud data, and generate a digital elevation model (DEM). Combined with the stratum depth information in the drilling data, establish an underground stratification model.

[0188] Specifically, the borehole data provides the lithology boundary of each stratum, and the point cloud surface model provides the geometric reference of the surface. The two are connected and integrated in the coordinate system to form a unified 3D geological model of the surface and underground. This model can fully express the geometric form of the surface and the underground geological structure, providing a reliable geometric and geological basis for engineering planning and analysis.

[0189] S2063. Add necessary geological attribute information to the generated BIM model to distinguish the excavation volume of different lithologies and strata.

[0190] First, the lithology attributes of the strata (such as soil, sandstone, shale, etc.) are bound to each layer of the model to ensure that each layer has a clear lithology definition. Secondly, combined with the geological exploration report, physical parameters related to excavation engineering requirements (such as density, volume, permeability, etc.) are added. These attribute information can intuitively reflect the engineering characteristics of different strata, ensuring that the BIM model has a high degree of accuracy and practicality in subsequent engineering quantity calculation and analysis.

[0191] S2064. Output a complete geological BIM model before excavation for subsequent engineering quantity calculation and construction planning.

[0192] The final model not only contains geometric information of the surface and underground, but also integrates geological attributes and engineering parameters, which can provide comprehensive data support for excavation projects. The model has clear surface features and accurate underground lithology distribution, and can be directly used for detailed calculation of engineering quantities, construction simulation and engineering decision-making. Through the application of geological BIM models, the planning efficiency and construction management level of geological engineering can be significantly improved.

[0193] S3. After the excavation work is completed, the oblique photography technology is used to obtain the point cloud data of the same area, and the geological model is constructed again. In view of the unclear and sparse point cloud recognition in the slope foot area, the inverse distance weighted interpolation algorithm is used to optimize the data accuracy and construct the post-excavation geological BIM model.

[0194] In the embodiment of the present application, after the excavation work is completed, the oblique photography technology is used to collect data from the excavation area again, obtain updated point cloud data, and re-construct the geological model. In addition, in order to solve the problem of unclear and sparse point cloud data in the slope foot area, the inverse distance weighted interpolation algorithm is used to optimize the point cloud data. Finally, the optimized point cloud data is combined with the pre-excavation geological BIM model to reconstruct the post-excavation geological BIM model.

[0195] The final post-excavation geological BIM model can accurately display the three-dimensional terrain changes after excavation, including regional features such as slopes, slope toes and faults, and contains high-precision surface and underground information. Through the dynamic update of the model, the changes before and after excavation can be compared in real time, providing reliable support for engineering construction management, excavation effect evaluation and geological analysis.

[0196] Specifically, step S3 includes the following steps:

[0197] S301. After the excavation work is completed, the drone oblique photography technology is used again to collect multi-view images of the same area, including vertical images and oblique images, to ensure that the image data covers the key features of the entire excavation area. During the collection process, focus on the slope foot, slope surface and other complex geological areas to ensure high resolution and comprehensive coverage of the image data. By importing the multi-view image data into the modeling software (such as PIX4D), after image stitching and reconstruction processing, high-precision three-dimensional point cloud data of the excavated area is generated.

[0198] The detailed process can be referred to steps S101-S103, and this application will not repeat it.

[0199] S302. Based on the three-dimensional point cloud data of the excavated geological body obtained in step S101, for areas with sparse point clouds, the inverse distance weighted interpolation algorithm is used to complete the elevation values ​​of missing points. In view of the problem of unclear point cloud recognition in the toe area of ​​the slope, the inverse distance interpolation combined with the local trend fitting method is used to adjust the interpolation strategy and parameters.

[0200] Furthermore, the specific operation of step S302 includes the following detailed process:

[0201] S3021. Perform density analysis on the three-dimensional point cloud data of the excavated geological body to identify sparse areas.

[0202] For each point p in the point cloud data i , based on the number of points in its neighborhood |N i |To perform statistical analysis:

[0203] |N i |={p j ∣∥p i -p j ∥≤r}

[0204] Among them, |N i | indicates point p i The number of points in the neighborhood, r is the neighborhood radius set by the user, usually 0.5 meters, for each point p i If the number of points in its neighborhood is less than the preset number of points, then p i The area is a sparse area. This analysis ensures the accurate identification of sparse areas and provides target areas for subsequent interpolation processing.

[0205] In one possible embodiment, if |N i | is less than 10, then p i The area is a sparse area.

[0206] 3022. For sparse areas, the embodiment of the present application uses the inverse distance weighted interpolation algorithm (IDW) to complete the elevation value Z (x, y) of the missing point, and the expression is:

[0207]

[0208] In the formula, Z i is the neighborhood point p i The known elevation value, d i is the distance between the interpolation point and the neighborhood point, and p is the distance weight index, which is 2. Through this algorithm, the point cloud in the sparse area is continuously completed to ensure that the generated elevation model has a high accuracy.

[0209] 3023. From the point cloud data of the slope foot area classified in step S2033, mark the key areas where the features are not clearly expressed, especially the boundary points or high curvature points, and extract and mark these areas to support subsequent trend analysis and interpolation optimization. These feature points contain the geometric characteristics of the area boundary and are key data points that reflect important geometric changes in the slope foot area.

[0210] 3024. In the marked slope foot area, select the feature points (x i ,y i ,z i ), and the local trend equation is established by using the linear regression fitting method, the formula is:

[0211] z=ax+by+c

[0212] Among them, a, b, and c are fitting parameters. The linear trend equation can provide the overall height change trend of the region, provide additional reference information for the interpolation algorithm, and improve the interpolation accuracy.

[0213] 3025. Combined with the local trend fitting results, the parameters of the inverse distance weighted interpolation algorithm are optimized. In sparse areas, the interpolation radius r is adjusted to 0.3 meters, and the number of interpolation neighborhood points is increased to 20. At the same time, the trend fitting results are combined with the IDW interpolation results to generate the final interpolation elevation value according to the weight.

[0214] In the embodiment of the present application, the elevation values ​​of trend fitting and the elevation values ​​of inverse distance weighted interpolation are assigned weights of 0.6 and 0.4 respectively. This step combines the trend fitting and interpolation results to improve the expression ability of the boundary features of sparse areas, especially significantly optimizing the model accuracy in the slope foot area.

[0215] 3026. After the interpolation is completed, the neighborhood mean smoothing method is used to optimize the interpolation area, thereby eliminating the abruptness of the interpolation area, reducing the discreteness between the interpolation points, and making the terrain transition more natural. The smoothing formula is:

[0216]

[0217] in, The smoothed point position, p j Points in the neighborhood, |N i | is the number of neighborhood points. The mean smoothing method can significantly reduce the abruptness caused by calculation errors in the interpolation area, making the optimized terrain smoother and more continuous.

[0218] 3027. After completing interpolation and optimization, compare the elevation values ​​of the interpolation area with the original three-dimensional point cloud data of the excavated geological body, and verify the accuracy of the interpolation area by calculating the mean square error (MSE). The calculation formula for the mean square error is:

[0219]

[0220] in, is the elevation value of the interpolation point, is the actual elevation value in the original point cloud data, and n is the number of verification points. By calculating the mean square error, it can be ensured that the accuracy of the interpolation optimization area meets the requirements of geological modeling, providing reliable elevation data support for the construction of subsequent geological BIM models.

[0221] S303, the optimized point cloud data is merged with the drilling data before excavation, and the surface point cloud and underground drilling information are integrated through a unified coordinate system. The point cloud data provides the latest geometric shape of the surface after excavation, and the drilling data reflects the lithology distribution and depth information of the underground strata.

[0222] Using Revit modeling software, point cloud data is used to generate a three-dimensional surface model, including the slope surface, slope foot and other key areas after excavation; at the same time, the underground geological layering model is constructed in combination with the drilling data to fully express the geological characteristics of the surface and underground in the same model. By combining the geometric information of the model with the geological attribute information, the accurate expression of the geological characteristics after excavation can be achieved.

[0223] The resulting post-excavation geological BIM model can not only intuitively display the three-dimensional morphological changes of the excavation area, but also accurately reflect the geological characteristics and their attribute information. This model provides high-precision data support for construction management, excavation quantity calculation, and geological change analysis, helping project planning and construction decision-making to achieve more efficient and scientific management.

[0224] S4. The geological BIM model before excavation is divided into regions according to the lithology, and the elevation difference of the geological BIM model before and after excavation is calculated pixel by pixel using the robust elevation difference calculation method based on the neighborhood window for the digital elevation models of the two stages before and after excavation.

[0225] In the embodiment of the present application, the geological BIM model before excavation is divided into regions according to the lithology information, and the model is divided into different lithology regions by combining the lithology attributes in the drilling data (such as soil layer, sandstone, shale, etc.) and the spatial geometric characteristics in the model. The division of each region is based on geological attributes and spatial distribution characteristics to ensure that the model can accurately reflect the layered structure and lithology boundaries of the geological body and provide accurate basic data for subsequent analysis. Then, for the digital elevation model before and after excavation, the robust elevation difference calculation method based on the neighborhood window is used to calculate the pixel-by-pixel elevation difference of the two-stage model.

[0226] By calculating the elevation difference pixel by pixel, we can generate an elevation change map of the excavation area, clearly showing the spatial distribution characteristics of the excavation volume. This step provides accurate basic data support for subsequent excavation volume calculation and geological change analysis, and also provides an important reference for construction planning and decision optimization.

[0227] Specifically, step S4 also includes:

[0228] S401. Marking different lithology areas based on the lithology-differentiated three-dimensional point cloud data before excavation.

[0229] In the embodiment of the present application, based on the lithology classification result of step S204, different lithology areas in the geological BIM model are labeled. By combining the drilling exploration data with the point cloud classification results, clear labels are given to each lithology area (such as sandstone, shale, soil layer, etc.) in the model. These annotations not only distinguish different geological areas, but also provide an accurate classification basis for subsequent elevation difference calculations and excavation engineering quantity statistics, ensuring that the elevation difference calculation results can be classified and summarized by lithology, providing data support for analyzing the excavation characteristics of each lithology area.

[0230] S402. Based on the marked lithological area, using the elevation data generated before and after excavation, the neighborhood window method is used to calculate the elevation difference pixel by pixel. For each pixel, the elevation data before and after the excavation are compared within the neighborhood window, and the minimum absolute value is selected as the true elevation difference. At the same time, the elevation difference threshold is set to mark the change point or the unchanged point, and the pixel-level result is mapped to the lithological area.

[0231] In the embodiment of the present application, based on the marked lithological area, the digital elevation model (DEM) generated before and after excavation is used to calculate the elevation difference pixel by pixel using the neighborhood window method. For each pixel point, a neighborhood window is defined, and the elevation data before and after excavation are compared within the window range. The minimum absolute value of the elevation difference in the neighborhood is calculated and used as the true elevation difference of the pixel.

[0232] This method can avoid erroneous elevation difference calculations caused by noise or outliers. At the same time, the elevation difference threshold is set to mark the changed points and unchanged points. For example, pixels whose elevation difference exceeds the threshold are judged to have changed, otherwise they are marked as unchanged. Finally, the pixel-level elevation difference results are mapped back to the lithology area to form an elevation change detection result based on lithology classification, realizing high-precision geological change detection. This step can clearly reflect the elevation change characteristics of different lithology areas and provide reliable data support for excavation volume calculation and geological analysis.

[0233] Furthermore, the specific operation of step S402 includes the following detailed process:

[0234] S4021, setting an appropriate neighborhood window size according to the geological body characteristics, selecting the window center pixel, locating its coordinates in the post-excavation stage elevation data, and extracting all pixel values ​​corresponding to the window range in the pre-excavation stage elevation data;

[0235] Specifically, according to the local change characteristics of the geological body, an appropriate neighborhood window size (2w+1) is set, and the window side length is w=3, corresponding to a 7×7 grid range. When selecting the window size, it is necessary to ensure that the local characteristic changes of the geological body can be covered while avoiding the interference of too much irrelevant data. For each window center pixel (i, j), its coordinates are located in the elevation data of the post-excavation stage, and all pixel values ​​corresponding to the window range in the elevation data of the pre-excavation stage are extracted. This process provides data support for the calculation of pixel-by-pixel elevation differences while ensuring the accurate capture of local changes.

[0236] S4022. For each pixel within the window range, calculate the elevation difference between the center pixel (i, j) of the window after excavation and the neighboring pixel (p, q) before excavation. Arter (i,j)-d Before (p,q)|. Compare all elevation differences one by one in the window, and take the minimum absolute value as the true elevation difference d of the current pixel D (i,j):

[0237] d D (i,j)=min{|d After (i,j)-d Before (p,q)|}

[0238] Among them, d D (i,j) is the elevation difference of pixel (i,j), d After (i,j) is the elevation value of the center pixel in the DEM at the later stage, d Before(p,q) is the elevation value of the pixel (p,q) in the window in the previous stage DEM, and (p,q) is the pixel coordinate in the window. This method can effectively reduce the interference of noise or abnormal data and obtain a more robust elevation difference.

[0239] S4023, by setting the elevation difference threshold T Height (The value is 0.5 meters) to mark the elevation difference of each pixel. If the actual elevation difference of the current pixel is greater than or equal to the elevation difference threshold, it is marked as a change point, otherwise it is marked as an unchanged point. The formula for the change marking rule is as follows:

[0240]

[0241] This marking method can intuitively distinguish the changed points and unchanged points in the excavated area, providing a basis for the subsequent statistical analysis of the lithology area.

[0242] S4024. Count the proportion of change points in each lithology area, and calculate the proportion of change pixels to total pixels. If the proportion of change points in a certain lithology area reaches a preset threshold (for example, 30%), the area is marked as a change area; if the proportion of change points is lower than the threshold, it is marked as an unchanged area. For unchanged areas, they will no longer participate in subsequent excavation volume calculations. In this way, the amount of calculation can be effectively reduced, while focusing on areas with significant changes, improving analysis efficiency and accuracy.

[0243] Through the above steps, the pixel-by-pixel elevation difference calculation and the change marking of the lithology area are completed, providing high-precision basic data support for the subsequent excavation volume calculation and change area analysis.

[0244] S5. Based on the elevation difference calculation results, calculate the volume change of each lithology area and obtain the engineering volume of the entire excavation area.

[0245] In order to accurately calculate the volume changes within the excavation area, this application combines the elevation difference results to accurately partition each lithology area, and gradually calculates the volume changes of each area based on regular grid units, and finally statistics the total engineering volume of the entire excavation area.

[0246] The specific steps of step S5 are as follows:

[0247] S501. Divide each lithology area into regular grid units according to the elevation difference calculation result.

[0248] In the embodiment of the present application, according to the elevation difference calculation results of steps S401 and S402, different lithology areas (such as sandstone, shale, etc.) are divided into uniform grid units, and the side length a of the grid unit is defined.

[0249] The grid size needs to be set according to the complexity of the terrain and the calculation accuracy requirements, and this application does not make specific restrictions on this. For example, for areas with more complex terrain, the grid side length can be selected to be a smaller value (such as 0.5 meters) to more carefully capture the terrain change characteristics; for flatter areas, the grid side length can be appropriately increased to simplify the calculation. The main purpose of grid division is to discretize continuous terrain data to facilitate the calculation of volume changes unit by unit, while ensuring that the complexity of the terrain in the area is fully expressed.

[0250] S502: For each grid unit, calculate the volume change of the single grid based on the elevation difference and the grid area.

[0251] In the embodiment of the present application, for each grid unit, the elevation difference d is used. D (x,y) and the area of ​​the grid cell A grid Calculate the volume change of a single grid. The volume calculation formula is:

[0252] V grid =A grid ·d D (x,y)

[0253] Among them, V grid Represents the volume change of a single grid cell, V grid =a 2 is the area of ​​the grid cell, d D (x, y) is the elevation difference of the center point of the grid unit. If the elevation difference is a negative value, it means that the grid unit has been excavated; if the elevation difference is a positive value, it means that the grid unit has been piled up. By calculating the grid by grid, the distribution characteristics of the volume change in the area can be accurately captured.

[0254] S503, accumulate the volume changes of all grid cells in each lithology area to calculate the total volume change of the area. The accumulation formula is:

[0255]

[0256] Among them, V rocktype It represents the total volume change of a certain lithology area (such as sandstone, clay), and n is the total number of grid cells in the area. For grid cells with negative elevation difference, their volume change is counted as excavation volume; for grid cells with positive elevation difference, their volume change is counted as accumulation volume.

[0257] This regional cumulative calculation method can be used to refine the volume changes in each lithological area, helping to analyze the specific contributions of different geological areas in excavation projects. At the same time, it provides accurate volume change results to support subsequent excavation management and analysis.

[0258] S504, summarizing the volume changes of all lithology areas and calculating the total engineering quantity of the entire excavation area. The final total excavation engineering quantity is obtained by accumulating the volume changes of each lithology area:

[0259]

[0260] Among them, V total It is the total engineering volume of the entire excavation area, including the sum of the excavation volume and the accumulation volume. The summarized data can fully reflect the volume changes of the entire excavation project, which is convenient for the overall evaluation of the engineering volume and the planning and management of excavation construction.

[0261] Through the above-mentioned step-by-step calculation, the volume change can be accurately refined to the grid unit, gradually accumulated to the rock type area, and finally form the total engineering quantity statistics of the entire excavation area. This engineering quantity assessment method based on elevation difference calculation not only ensures the accuracy of the calculation results, but also can show the specific change characteristics of different rock type areas in detail, providing comprehensive data support for construction cost estimation, resource allocation and engineering schedule arrangement.

[0262] In order to implement the above embodiments, the present application also proposes a geological BIM modeling and engineering quantity calculation device based on point cloud sparse optimization. Figure 2 The present invention provides a structural diagram of a geological BIM modeling and engineering quantity calculation device based on point cloud sparse optimization. Figure 2 As shown, the device comprises:

[0263] The data acquisition module 100 is used to obtain geological body image data and three-dimensional point cloud data by using oblique photography technology, and to associate and combine the three-dimensional point cloud data with the drilling exploration data;

[0264] The pre-excavation geological model construction module 200 is used to perform data preprocessing on the three-dimensional point cloud data, and to construct a pre-excavation geological BIM model using the preprocessed three-dimensional point cloud data, image data and drilling exploration data. The data preprocessing includes denoising, classification, segmentation, lithology differentiation and registration.

[0265] The post-excavation geological model reconstruction module 300 is used to obtain point cloud data of the same area by using oblique photography technology after the excavation work is completed, and to re-construct the geological model. In view of the unclear and sparse point cloud recognition problems in the slope foot area, the inverse distance weighted interpolation algorithm is used to optimize the data accuracy and construct the post-excavation geological BIM model.

[0266] The elevation difference calculation module 400 is used to divide the geological BIM model before excavation into regions according to the lithology, and to calculate the elevation difference of the geological BIM model before and after excavation pixel by pixel using a robust elevation difference calculation method based on a neighborhood window for the digital elevation models of the two stages before and after excavation;

[0267] The engineering quantity calculation module 500 is used to calculate the volume change of each lithology area according to the elevation difference calculation result to obtain the engineering quantity of the entire excavation area.

[0268] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0269] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0270] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0271] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0272] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this application can be executed in parallel, sequentially or in different orders, as long as the expected results of the technical solution of this application can be achieved, and this document is not limited here.

[0273] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A geological BIM modeling and engineering quantity calculation method based on point cloud sparse optimization, characterized in that: The following steps are involved: Obtaining geological body image data and three-dimensional point cloud data using oblique photography technology, and associating and combining the three-dimensional point cloud data with drilling exploration data; Preprocessing the three-dimensional point cloud data, and constructing a geological BIM model before excavation using the preprocessed three-dimensional point cloud data, image data and drilling exploration data, wherein the data preprocessing includes denoising, classification, segmentation, lithology differentiation and registration; After the excavation work is completed, the oblique photography technology is used to obtain the point cloud data of the same area, and the geological model is constructed again. In view of the unclear and sparse point cloud recognition in the slope foot area, the inverse distance weighted interpolation algorithm is used to optimize the data accuracy and construct the post-excavation geological BIM model; The geological BIM model before excavation is divided into regions according to the lithology, and the pixel-by-pixel elevation difference calculation method based on the robust elevation difference calculation method based on the neighborhood window is used to calculate the geological BIM model before and after excavation for the digital elevation models of the two stages before and after excavation. Based on the elevation difference calculation results, the volume change of each lithology area is calculated to obtain the engineering volume of the entire excavation area.

2. The method according to claim 1, characterized in that The method of obtaining geological body image data and three-dimensional point cloud data by using oblique photography technology, and associating and combining the three-dimensional point cloud data with drilling exploration data, includes: Collect multi-view image data of geological bodies by using drone oblique photography technology, wherein the multi-view image data includes vertical images and oblique images; Based on PIX4D software, the collected multi-view image data is imported and processed, dense point cloud data is generated through the software's automated workflow, the spatial geometric information of the geological body is extracted, and three-dimensional point cloud data containing high-precision three-dimensional coordinates is formed; The drilling exploration data including drilling location, formation depth, lithology distribution and physical properties are collected and integrated to establish a geological body internal structure information database and associate it with the three-dimensional point cloud data.

3. The method according to claim 2, characterized in that The preprocessing of the three-dimensional point cloud data, and construction of a pre-excavation geological BIM model using the preprocessed three-dimensional point cloud data, image data and drilling exploration data, includes: De-noising the three-dimensional point cloud data, using a statistical filtering algorithm to remove isolated points, noise points and irregular point groups to ensure the smoothness and accuracy of the point cloud data; Based on a multi-layer dynamic graph convolutional neural network, the denoised point cloud data is classified, and geological features, vegetation, buildings and noise in the point cloud are extracted according to geometric features, providing a labeling and classification basis for the subsequent elimination of non-geological feature point clouds; Based on the classified three-dimensional point cloud data, deep feature learning is used to segment the data, extract feature areas related to the geological body, and eliminate point cloud data unrelated to the geological body; By distinguishing the lithology of the pure geological point cloud data generated after segmentation, and combining the drilling exploration data and deep learning classification model, the point cloud data of the geological body is refined into different lithology areas; The lithologically differentiated three-dimensional point cloud data is registered, and the three-dimensional point cloud data scanned multiple times is spatially aligned by an iterative closest point algorithm, and the coordinates are unified and integrated with the drilling exploration data to ensure that the point cloud and image data and geological drilling information are unified in the same coordinate system, eliminate spatial errors and integrate surface and underground information; The pre-processed three-dimensional point cloud data, image data and drilling exploration data are used to construct a complete pre-excavation geological BIM model to express the three-dimensional shape, internal structure and attribute information of the geological body.

4. The method according to claim 3, characterized in that The three-dimensional point cloud data, image data and drilling exploration data completed by pre-processing are used to construct a complete geological BIM model before excavation, which is used to express the three-dimensional shape, internal structure and attribute information of the geological body, including: Importing the pre-processed three-dimensional point cloud data and drilling exploration data into BIM modeling software; The geometric features of the ground surface, including the slope surface, slope foot and fault area, are extracted through the three-dimensional point cloud data, and a digital elevation model DEM of the ground surface is generated; an underground layered model is established in combination with the drilling exploration data, and the stratum depth information in the borehole is connected with the point cloud surface model to form a unified three-dimensional geological BIM model of the ground surface and the underground; Add necessary geological attributes to the generated BIM model to distinguish the excavation volume of different lithologies and strata. The geological attributes include density and volume parameters related to the lithology attributes of the strata and the excavation engineering requirements; Output a complete geological BIM model before excavation.

5. The method according to claim 4, characterized in that After the excavation work is completed, the oblique photography technology is used to obtain the point cloud data of the same area, and the geological model is constructed again. In view of the unclear and sparse point cloud recognition in the slope foot area, the inverse distance weighted interpolation algorithm is used to optimize the data accuracy and construct the post-excavation geological BIM model, including: After the excavation work is completed, the UAV oblique photography technology is used again to collect multi-view images of the same area, and the three-dimensional point cloud data of the geological body after excavation is obtained by using the multi-view image collection data collected again; Based on the 3D point cloud data of the excavated geological body, the inverse distance weighted interpolation algorithm is used to complete the elevation values ​​of the missing points in the sparse point cloud areas. In order to solve the problem of unclear and sparse point cloud recognition in the slope foot area, the inverse distance interpolation combined with the local trend fitting method is used to adjust the interpolation strategy and parameters. The optimized 3D point cloud data was combined with the drilling exploration data before excavation, and the post-excavation geological BIM model was constructed using Revit modeling software.

6. The method according to claim 5, characterized in that The three-dimensional point cloud data based on the excavated geological body is used to fill in the elevation values ​​of the missing points in the area with sparse point clouds. In order to solve the problem of unclear and sparse point cloud recognition in the foot of the slope area, the inverse distance interpolation is combined with the local trend fitting method to adjust the interpolation strategy and parameters, including: Density analysis is performed on the 3D point cloud data of the excavated geological body to identify sparse areas. For each point p i If the number of points in its neighborhood is less than the preset number of points, then p i The area is a sparse area; For sparse areas, the inverse distance weighted interpolation algorithm is used to complete the elevation value Z(x,y) of the missing point. The expression is: In the formula, Z i is the neighborhood point p i The known elevation value, d i is the distance between the interpolation point and the neighborhood point, and p is the distance weight index; From the point cloud data of the slope foot area obtained based on classification before excavation, mark the key areas where sparse or unclear features are expressed; In the marked slope foot area, feature points in the sparse area neighborhood are selected, and the local trend equation is established using the linear regression fitting method; According to the fitting trend result obtained after solving the local trend equation, the parameters of the inverse distance interpolation are optimized, and the preliminary elevation value of the trend fitting is combined with the interpolation result of the inverse distance weighted algorithm to generate the final interpolation elevation value according to the weight; After the interpolation is completed, the neighborhood mean smoothing method is used to eliminate the abruptness of the interpolation area; After completing the interpolation and optimization, the elevation values ​​of the interpolation area are compared with the original three-dimensional point cloud data of the geological body after excavation, and the accuracy of the interpolation area is verified by calculating the mean square error.

7. The method according to claim 6, characterized in that The geological BIM model before excavation is divided into regions according to lithology, and the digital elevation models before and after excavation are subjected to pixel-by-pixel elevation difference calculation using a robust elevation difference calculation method based on a neighborhood window, including: Marking different lithology areas according to the three-dimensional point cloud data before excavation that has been differentiated by lithology; Based on the annotated lithological areas, the elevation data generated before and after excavation are used to calculate the elevation difference pixel by pixel using the neighborhood window method. For each pixel, the elevation data before and after are compared within the neighborhood window, and the minimum absolute value is selected as the true elevation difference. At the same time, the elevation difference threshold is set to mark the change points or unchanged points, and the pixel-level results are mapped to the lithological areas.

8. The method according to claim 7, characterized in that Based on the marked lithological area, the elevation data generated before and after excavation are used to calculate the elevation difference pixel by pixel using the neighborhood window method. For each pixel, the elevation data before and after the excavation are compared in the neighborhood window, and the minimum absolute value is selected as the true elevation difference. At the same time, the elevation difference threshold is set to mark the change point or the unchanged point, and the pixel-level result is mapped to the lithological area, including According to the characteristics of the geological body, the appropriate neighborhood window size is set, the center pixel of the window is selected, its coordinates are located in the elevation data of the post-excavation stage, and all pixel values ​​corresponding to the window range in the elevation data of the pre-excavation stage are extracted; For each pixel in the window, calculate the elevation difference between the central pixel after excavation and the neighboring pixels before excavation, and take the minimum absolute value as the true elevation difference of the current pixel; Each pixel is marked as changed by setting an elevation difference threshold. If the actual elevation difference of the current pixel is greater than or equal to the elevation difference threshold, it is marked as a changed point, otherwise it is marked as an unchanged point. The proportion of change points in each lithology area is counted, and the proportion of change pixels to total pixels is calculated. If the proportion of change points in a lithology area reaches a preset threshold, the area is marked as a change area, otherwise the area is marked as an unchanged area.

9. The method according to claim 8, characterized in that The above-mentioned method calculates the volume change of each lithology area based on the elevation difference calculation result to obtain the engineering quantity of the entire excavation area, including: Based on the elevation difference calculation results, each lithological area is divided into regular grid cells; For each grid cell, the volume change of a single grid is calculated based on the elevation difference and the grid area; The volume changes of all grids in each lithology area are accumulated to obtain the volume change of the area. For the negative elevation difference area, the volume is counted as the excavation volume; for the positive elevation difference area, the volume is counted as the accumulation volume. The volume change data of all lithology areas are summarized and the total engineering volume of the entire excavation area is calculated.

10. A geological BIM modeling and engineering quantity calculation device based on point cloud sparse optimization, characterized in that: include: A data acquisition module is used to acquire geological body image data and three-dimensional point cloud data using oblique photography technology, and to associate and combine the three-dimensional point cloud data with drilling exploration data; A pre-excavation geological model construction module is used to perform data preprocessing on the three-dimensional point cloud data, and to construct a pre-excavation geological BIM model using the preprocessed three-dimensional point cloud data, image data and drilling exploration data. The data preprocessing includes denoising, classification, segmentation, lithology differentiation and registration. The post-excavation geological model reconstruction module is used to obtain point cloud data of the same area using oblique photography technology after the excavation work is completed, and to re-construct the geological model. In view of the unclear and sparse point cloud recognition in the slope foot area, the inverse distance weighted interpolation algorithm is used to optimize the data accuracy and construct the post-excavation geological BIM model. The elevation difference calculation module is used to divide the geological BIM model before excavation into regions according to the lithology, and to calculate the pixel-by-pixel elevation difference of the geological BIM model before and after excavation using a robust elevation difference calculation method based on a neighborhood window for the digital elevation models of the two stages before and after excavation. The engineering quantity calculation module is used to calculate the volume change of each lithology area based on the elevation difference calculation results to obtain the engineering quantity of the entire excavation area.

Images