Three-dimensional dynamic visualization simulation method and system for landslide disaster

Abstract

The application provides a three-dimensional dynamic visualization simulation method and system for landslide disasters, comprising: acquiring multi-source data of a landslide area; identifying static points and dynamic sliding points in the landslide area based on the multi-source data; constructing a three-dimensional geometric model of the landslide area according to the multi-source data, a static point index list and a dynamic sliding point set, and rendering the static points and the dynamic sliding points with differentiated colors and transparencies; and realizing dynamic visualization simulation of the whole process of the landslide from starting, flowing to piling up in a real three-dimensional terrain environment through time sequence control and dynamic rendering. The application realizes full-process automatic processing, cross-platform data adaptation and dynamic simulation in a real geographic environment by integrating a three-dimensional geographic information system, real-time dynamic simulation and high-performance rendering, and provides reliable and intuitive technical support for mine safety management, geological disaster prediction and emergency decision-making.

Description

Technical Field

[0001] This application belongs to the field of landslide prediction and simulation technology, and in particular relates to a three-dimensional dynamic visualization simulation method and system for landslide disasters. Background Technology

[0002] With the continuous expansion of non-coal mining, factors such as surface disturbance, changes in soil and rock structure, and rainfall infiltration have significantly increased the frequency of geological disasters such as landslides, collapses, and debris flows. Landslide disasters are characterized by their suddenness, wide impact range, rapid evolution, and complex disaster chains. Once they occur, they not only cause mine production interruptions and equipment damage, but in more serious cases, they can cause casualties and even trigger secondary disasters (such as river blockage and dam failure), resulting in long-term impacts on regional ecological security. Therefore, dynamic simulation of landslide disasters, especially simulations that accurately correspond to the real geographical environment, has become an important technical direction for mine safety supervision and disaster prevention and mitigation.

[0003] However, the landslide risk monitoring and simulation methods commonly used in the industry currently rely mainly on two-dimensional geographic information systems (GIS) or static engineering analysis models, which have problems such as limited display, insufficient simulation capabilities, and insufficient accuracy.

[0004] Therefore, there is an urgent need for a new mine safety management and geological disaster visualization technology to address the shortcomings of the existing technologies.

[0005] The foregoing statements are for informational purposes only and are not intended to provide background information in connection with this application. Unless otherwise stated herein, the content described in this section is not prior art to the rest of this application. Summary of the Invention

[0006] This invention proposes a three-dimensional dynamic visualization simulation method and system for landslide disasters. By integrating a three-dimensional geographic information system, real-time dynamic simulation, and high-performance rendering, it achieves fully automated processing, cross-platform data adaptation, and dynamic simulation in a real geographical environment, providing reliable and intuitive technical support for mine safety management, geological disaster prediction, and emergency decision-making. It solves key technical problems in visualization, accuracy, and performance of traditional landslide simulation.

[0007] Furthermore, this application utilizes the Web-based Cesium platform, employing technologies such as cross-platform data adaptation, high-frequency data processing, and differentiation between static points and dynamic sliding bodies. It also employs a frame-skipping downsampling strategy for high-frequency data, achieving the following technical effects: 1) Through a cross-platform adaptation mechanism, simulation data from different sources can be displayed in a unified Web 3D platform, lowering the barrier to entry and improving the system's versatility and ease of use. 2) Through a series of data preprocessing and rendering optimizations, large-scale, high-frequency landslide dynamic simulations can run smoothly in ordinary browsers, facilitating sharing and decision support; and high-performance web-based visualization has been achieved.

[0008] 3) Through high-precision spatial coordinate mapping and differentiated rendering of static / dynamic areas, the simulation results strictly correspond to the real terrain and can intuitively display the impact range of landslides, assisting in accurate analysis; thus enhancing the geospatial realism and analytical capabilities of the simulation results.

[0009] According to a first aspect of the embodiments of this application, a three-dimensional dynamic visualization simulation method for landslide disasters is provided, comprising the following steps: Acquire multi-source data of the landslide area, including digital elevation model (DEM) data, geological structure data, gridded topographic information, and physical parameters of the landslide body; Based on multi-source data, static points and dynamic sliding points in the landslide area are identified, and a static point index list and a dynamic sliding point set are generated. Based on multi-source data, a static point index list, and a dynamic sliding point set, a three-dimensional geometric model of the landslide area is constructed, and the static points and dynamic sliding points are rendered with differentiated colors and transparency. By loading the 3D geometric model into a Cesium-based 3D visualization platform, and through time-series control and dynamic rendering, the entire process of a landslide, from initiation and flow to deposition, is dynamically visualized and simulated in a real 3D terrain environment.

[0010] In some embodiments of this application, the multi-source data also includes simulation output data from external mechanical simulation software; The method also includes: parsing, reconstructing, and mapping the simulation output data into standardized data through cross-platform data conversion and adaptation mechanisms. This enables seamless integration and recalculation of mechanical simulation results in a 3D visualization platform.

[0011] In some embodiments of this application, the method further includes: Multi-source data are preprocessed by converting the original non-spatial coordinate landslide observation data into standardized data with geospatial coordinates through data digitization, key control point mapping, and spatial coordinate transformation.

[0012] In some embodiments of this application, the original non-spatial coordinate landslide observation data is converted into standardized data with geospatial coordinates through data digitization, key control point mapping, and spatial coordinate transformation, including: The landslide observation data is unified to a local reference frame with the centroid of the dataset as the origin; The starting point, ending point, and intermediate control point of the landslide were selected as key geographical reference points, and the correspondence between their local coordinates and actual geographical coordinates was established. Based on the correspondence, the spatial transformation parameters are solved by weighted least squares method, and all original data points are mapped to the real geographic coordinate system with high precision.

[0013] In some embodiments of this application, preprocessing of multi-source data further includes: High-frequency landslide observation data are subjected to frame extraction and downsampling. Key frames are adaptively extracted from the original high-frequency time series data based on the rate of change of motion state. The rate of change of motion state is calculated based on the combined changes in the velocity of the center of mass of the landslide body and the changes in kinetic energy between adjacent frames. The entire process of data reading, coordinate mapping, frame extraction, and format conversion is automated through scripts.

[0014] In some embodiments of this application, identifying static points and dynamic sliding points in a landslide area includes: Traverse the coordinates of points in consecutive keyframes and compare the 3D coordinates of points with the same index in two adjacent frames. According to the determination rules, it is determined to be a static point, and its index is saved to the static point index list; Based on the static point index list, different color and transparency attributes are assigned to static points and dynamic sliding points respectively during rendering.

[0015] In some embodiments of this application, the steps of constructing a three-dimensional geometric model include: Convert geographic coordinates to Cartesian space coordinates to generate a vertex array; Based on the static point index list and the dynamic sliding point set, assign preset color and transparency values ​​to the static and dynamic points in the vertex array, respectively. A sequence of vertex arrays is used to generate a triangle index array, which is then used to construct a landslide surface mesh for rendering. The vertex array, color array, and index array are encapsulated into a Cesium.Geometry object, and a corresponding GeometryInstance instance is created for rendering.

[0016] According to a second aspect of the embodiments of this application, a three-dimensional dynamic visualization simulation system for landslide disasters is provided, comprising: The data acquisition and preprocessing module is used to acquire multi-source data of the landslide area, including digital elevation model (DEM) data, geological structure data, gridded topographic information, and physical parameters of the landslide body; The static point identification module is used to identify static points and dynamic sliding points in a landslide area based on multi-source data, and generate a static point index list and a dynamic sliding point set. The landslide geometry modeling module is used to construct a three-dimensional geometric model of the landslide area based on multi-source data, a static point index list, and a dynamic sliding point set, and to render static points and dynamic sliding points with differentiated colors and transparency. The dynamic rendering and interactive control module is used to load 3D geometric models into a Cesium-based 3D visualization platform. Through time-series control and dynamic rendering, it enables dynamic visualization simulation of the entire process of landslides from initiation, flow to deposition in a real 3D terrain environment.

[0017] According to a third aspect of the embodiments of this application, a three-dimensional dynamic visualization simulation device for landslide disasters is provided, comprising: a storage unit for storing executable instructions; and a processing unit for connecting to the storage unit to execute the executable instructions to complete a three-dimensional dynamic visualization simulation method for landslide disasters.

[0018] According to a fourth aspect of the embodiments of this application, a computer-readable storage medium is provided having a computer program stored thereon; the computer program is executed by a processor to implement a three-dimensional dynamic visualization simulation method for landslide disasters.

[0019] The three-dimensional dynamic visualization simulation method and system for landslide disasters proposed in this application includes acquiring multi-source data of the landslide area, including digital elevation model (DEM) data, geological structure data, gridded terrain information, and physical parameters of the landslide body; identifying static points and dynamic sliding points in the landslide area based on the multi-source data; constructing a three-dimensional geometric model of the landslide area based on the multi-source data, the static point index list, and the dynamic sliding point set, and rendering the static points and dynamic sliding points with differentiated colors and transparency; loading the three-dimensional geometric model into a Cesium-based three-dimensional visualization platform, and realizing dynamic visualization simulation of the entire process of landslide from initiation, flow to deposition in a real three-dimensional terrain environment through time series control and dynamic rendering.

[0020] This application integrates a 3D geographic information system, real-time dynamic simulation, and high-performance rendering to achieve fully automated processing, cross-platform data adaptation, and dynamic simulation in realistic geographic environments. This provides reliable and intuitive technical support for mine safety management, geological disaster prediction, and emergency decision-making. It solves key technical problems in visualization, accuracy, and performance of traditional landslide simulation.

[0021] Compared with existing technologies, it has the following differences: 1. Three-dimensional dynamic simulation and visualization; (1) Existing technologies mostly use two-dimensional maps or static three-dimensional models, which cannot intuitively display the spatial hierarchy and evolution process of landslides. However, this product uses time-series keyframes and Cesium.GeometryInstance instances to realize the dynamic display of landslides from initiation, flow to deposition, making the landslide evolution process visible and traceable.

[0022] (2) Problem solved: It solves the problem that traditional methods cannot dynamically present the entire process of landslides, enabling users to clearly understand the movement patterns of landslides and the distribution of dangerous areas.

[0023] 2. Distinguish between static points and dynamic sliding bodies; (1) Existing technologies typically treat the entire landslide area as a homogeneous body, lacking a distinction between stable and sliding areas. This product automatically identifies static and sliding points through continuous frame data comparison and displays them with different colors and transparency to dynamically reflect the range of landslide movement.

[0024] (2) Problem solved: It solves the problem that traditional methods cannot distinguish between stable and unstable areas within a landslide area, thereby improving the accuracy and reliability of landslide simulation.

[0025] 3. High-performance rendering and cross-platform support; (1) Traditional simulations rely on offline computing or inefficient rendering, making real-time display on the browser side impossible. This product utilizes Cesium GPU acceleration and asynchronous rendering to achieve real-time dynamic landslide simulation at millions of points. It supports cross-software and cross-platform data adaptation, seamlessly mapping external mechanical simulation results to the 3D visualization system.

[0026] (2) Problem solved: It solved the problems of insufficient visualization performance and data compatibility, and improved the user experience and decision-making reference value. Attached Figure Description

[0027] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 The diagram illustrates the steps of a three-dimensional dynamic visualization simulation method for landslide disasters according to an embodiment of this application. Figure 2 The diagram illustrates the data preprocessing steps according to an embodiment of this application; Figure 3 The diagram illustrates the steps of static point identification according to an embodiment of this application; Figure 4 The diagram illustrates the steps of constructing a three-dimensional geometric model according to an embodiment of this application; Figure 5 The diagram illustrates the overall flow of a three-dimensional dynamic visualization simulation method for landslide hazards according to an embodiment of this application. Figure 1 ; Figure 6 The diagram illustrates the overall flow of a three-dimensional dynamic visualization simulation method for landslide hazards according to an embodiment of this application. Figure 2 ; Figure 7 The diagram shows a dynamic simulation scenario of a landslide according to an embodiment of this application; Figure 8 The diagram shows a structural schematic of a three-dimensional dynamic visualization simulation system for landslide disasters according to an embodiment of this application; Figure 9 The diagram shows a structural schematic of a three-dimensional dynamic visualization simulation device for landslide disasters according to an embodiment of this application. Detailed Implementation

[0028] Regarding this application, with the continuous expansion of the scale of non-coal mining, the stability of mine slopes faces severe challenges, and geological disasters such as landslides, collapses and debris flows occur frequently, seriously threatening the safety of mine production and the safety of people's lives and property.

[0029] Currently, the field of geological disaster simulation and visualization is gradually evolving from traditional two-dimensional static analysis to three-dimensional dynamic display. Three-dimensional visualization technology can intuitively present complex topography, landslide area range, landslide flow path, and geological structure. Combined with monitoring data and elevation information, it can perform dynamic simulation, achieving accurate reproduction of the entire landslide process on real terrain, providing technical support for mine management and disaster early warning.

[0030] Currently, the commonly used methods for landslide risk monitoring and simulation in the industry mainly rely on two-dimensional geographic information systems (GIS) or static engineering analysis models. Although existing technologies have achieved certain results in landslide risk assessment and simulation, they still have the following shortcomings: 1. Significant limitations of two-dimensional display: Traditional GIS systems mostly use two-dimensional maps, which are difficult to represent topographic relief, sliding direction, and spatial diffusion range, and are also difficult to represent the spatial hierarchy and dynamic evolution process of landslides. For mining areas with complex terrain, two-dimensional images often cannot accurately represent the actual movement path and deposition morphology of landslides.

[0031] Furthermore, it lacks visualization of dynamic processes: traditional methods can only display landslide results, but cannot present the landslide occurrence, evolution, and stabilization process.

[0032] 2. Weak dynamic simulation capabilities: Most existing systems can only display landslide results, but cannot present the entire process of landslide occurrence, evolution to stabilization, and lack simulations that correspond to actual geographical coordinates.

[0033] 3. Insufficient accuracy of geographic coordinates and environment: Existing simulations are mostly based on local coordinates or simplified terrain. The landslide model is not completely aligned with the actual geographical environment of the mining area, which affects the simulation's realism and visualization effect.

[0034] 4. Limited decision support capabilities: The lack of 3D visualization tools that link with real geographic data makes it difficult for managers to identify risk areas and assess disaster trends in a timely manner, affecting scientific decision-making.

[0035] Therefore, there is an urgent need for a new type of landslide dynamic visualization technology that is based on three-dimensional geospatial space and can accurately map the real environment.

[0036] This application utilizes real terrain data, monitoring data, and physical landslide models to simulate the initiation, flow, diffusion, and deposition processes of landslides in three-dimensional space. Through dynamic visualization techniques, abstract landslide data is transformed into three-dimensional dynamic images, achieving a simulation effect that is "visible, controllable, and accurately calculated."

[0037] The main innovation of this invention compared to existing technologies lies in: (1) Three-dimensional real environment modeling: Based on high-precision DEM, topographic images and geological data, a real three-dimensional topography of the landslide area is constructed to ensure that the movement of the landslide body strictly follows the actual slope height, slope and geological conditions.

[0038] (2) Accurate mapping of geographic coordinates: By matching key control points and transforming spatial coordinates, the landslide observation data is accurately mapped to real geographic coordinates, so as to achieve a high degree of consistency between the model and the mining area environment.

[0039] (3) Full-process dynamic landslide simulation: Combining time series keyframes, geometric instance control and dynamic calculation, the dynamic simulation of landslide from initiation, flow to accumulation is realized, making landslide movement visible, interactive and analyzable in the three-dimensional GIS environment.

[0040] This patented technology not only realistically reproduces the entire landslide process, but also accurately corresponds to the actual terrain of the mining area, providing a reliable, intuitive, and efficient technical means for mine safety management, risk prediction, and geological disaster prevention and control.

[0041] It has the following technical effects: 1) Realistic and dynamic display: The landslide evolution process is continuously and smoothly visualized in three-dimensional space, which can clearly show the difference between the sliding area and the stationary area, and reflect the speed and movement trend through changes in color and transparency.

[0042] 2) High-precision spatial alignment: The landslide model strictly corresponds to the real terrain, slope and height, ensuring that the simulation results are highly consistent with the actual geographical location.

[0043] 3) High-performance data processing and rendering: Intelligent frame extraction strategy significantly reduces data volume, ensuring the smoothness and real-time performance of dynamic simulation.

[0044] GPU-accelerated rendering supports real-time drawing of hundreds of thousands of points, achieving a fluid sliding effect.

[0045] 4) Cross-platform and data compatibility: It connects the data chain between external mechanical simulation software and 3D visualization platform, supports multi-source data integration and cross-platform application, and improves the system's versatility and scalability.

[0046] 5) Scientific research and engineering application support: It can be used for landslide evolution analysis, geological disaster research, mine safety assessment, emergency drills and other scenarios.

[0047] It provides data-driven dynamic simulations to support landslide risk analysis from qualitative to quantitative perspectives.

[0048] 6) High reliability and repeatability: Through high-precision coordinate mapping, standardized output and automated processing, the simulation results are ensured to be repeatedly generated in different regions and datasets, adapting to the needs of complex scenarios.

[0049] To make the technical solutions and advantages of the embodiments of this application clearer, the exemplary embodiments of this application will be described in further detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not an exhaustive list of all embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other.

[0050] Example 1 Figure 1 The diagram illustrates the steps of a three-dimensional dynamic visualization simulation method for landslide disasters according to an embodiment of this application.

[0051] like Figure 1 As shown, a three-dimensional dynamic visualization simulation method for landslide disasters is provided, including the following steps: S1: Acquire multi-source data of the landslide area, including digital elevation model (DEM) data, geological structure data, gridded topographic information, and physical parameters of the landslide body; S2: Based on multi-source data, identify static points and dynamic sliding points in the landslide area, and generate a static point index list and a dynamic sliding point set; S3: Based on multi-source data, a static point index list, and a dynamic sliding point set, construct a three-dimensional geometric model of the landslide area, and render the static points and dynamic sliding points with differentiated colors and transparency. S4: Load the 3D geometric model into a Cesium-based 3D visualization platform. By controlling the time-series display and hiding of multiple geometric instances, i.e., through time-series control and dynamic rendering, realize the dynamic visualization simulation of the entire process of landslide from initiation, flow to deposition in a real 3D terrain environment.

[0052] This application proposes a 3D dynamic visualization simulation method for landslide disasters, based on the 3D spatial rendering capabilities of the Cesium platform. This method achieves realistic, continuous, and interactive reproduction of the landslide process through multi-source data fusion, geometric modeling, spatial coordinate transformation, time-series control, and dynamic rendering. Its core lies in constructing a complete methodological system and cross-platform data adaptation mechanism based on 3D geological disaster simulation, forming a closed-loop process from data acquisition to 3D dynamic display, realizing high-precision mapping and dynamic expression of geomechanical simulation results on the visualization platform.

[0053] Therefore, this application, by integrating a 3D geographic information system, real-time dynamic simulation, and high-performance rendering, achieves fully automated processing, cross-platform data adaptation, and dynamic simulation in a real geographic environment, providing reliable and intuitive technical support for mine safety management, geological disaster prediction, and emergency decision-making. It solves key technical problems in visualization, accuracy, and performance of traditional landslide simulation.

[0054] The multi-source data also includes simulation output data from external mechanical simulation software; the corresponding methods include: parsing, reconstructing, and mapping the simulation output data into standardized data through cross-platform data conversion and adaptation mechanisms. This enables seamless integration and recalculation of mechanical simulation results in a 3D visualization platform.

[0055] The "external mechanical simulation data" comes from professional mechanical software such as PFC3D, FLAC3D, or ABAQUS. The method also includes parsing, reconstructing, and standardizing the output format of the external mechanical simulation data through customized data bridging scripts to achieve seamless cross-platform data import.

[0056] Figure 2 The diagram illustrates the steps of data preprocessing according to an embodiment of this application.

[0057] like Figure 2 As shown, the method also includes: S5: Preprocess multi-source data by converting raw, non-spatial coordinate landslide observation data or external mechanical simulation data into standardized data with geographic spatial coordinates through data digitization, key control point mapping, and spatial coordinate transformation.

[0058] In practice, S5 transforms the original non-spatial coordinate landslide observation data into standardized data with geospatial coordinates through data digitization, key control point mapping, and spatial coordinate transformation. This includes: S51: Using the centroid of the dataset as the origin, the original non-spatial coordinate data of the landslide observation data are unified to a local reference frame; S52: Select the starting point, ending point and intermediate control point of the landslide as key geographical reference points, and establish the correspondence between their local coordinates and actual geographical coordinates; S53: Based on the correspondence, the spatial transformation parameters are solved by weighted least squares method, and all original data points are mapped to the real geographic coordinate system with high precision.

[0059] In preferred implementation, the example is mapping the local coordinates or raw observation data output by external simulation software to a real geographic coordinate system (such as the WGS84 coordinate system).

[0060] The weighted least squares mapping model based on control points is: P g = T * P L + Δ; Among them, P L = [X L Y L Z L ] T : These are the coordinates of a point in the original data (local coordinate system); P g = [lon, lat, elev] T Target geographic coordinates (longitude, latitude, elevation); T is a 3x3 spatial transformation matrix, which can be decomposed into rotation (R), scaling (S), and non-orthogonal deformation (D) matrices, i.e., T = R * S * D. This matrix is ​​solved by control point fitting.

[0061] Δ = [Δx, Δy, Δz] T , is the translation vector.

[0062] The solution process is as follows: Select n pairs (n≥4) of high-confidence control points { (P L (i), P g (i) | i=1...n}; The error equation is constructed as follows: E i = P g (i)- (T * P L (i) + Δ); Introducing the weight matrix W iThe weights can be assigned based on the positioning accuracy or importance of the control points (e.g., the landslide crown and shear outlet have higher weights). The transformation parameters T and Δ are solved using the weighted least squares method to minimize the total error, as shown in the following formula: min Σ (W i * || E i || 2 ).

[0063] In some embodiments of this application, preprocessing of multi-source data further includes: High-frequency landslide observation data are subjected to frame extraction and downsampling. Key frames are adaptively extracted from the original high-frequency time-series data based on the rate of change of motion state. The rate of change of motion state is calculated based on the combined changes in the velocity of the center of mass of the landslide body and the changes in kinetic energy between adjacent frames. It also uses automated scripts to achieve full-process batch processing of data reading, coordinate mapping, frame extraction, and format conversion.

[0064] In preferred implementation scenarios, the high-frequency time-series observation data (e.g., 60 frames / second) is downsampled to significantly reduce the data volume while ensuring dynamic continuity.

[0065] Based on the frame-jumping decision function of the motion energy change rate, the "motion state change rate" V of the k-th frame is defined. k : V k = α * || V k - V {k-1} || / Δt + β * || (E k - E {k-1} ) / E {k-1} || The parameters are defined as follows: V k : The average velocity vector of the center of mass of the sliding body in the kth frame; E k : The total kinetic energy of the sliding body in the kth frame (which can be estimated based on particle velocity); Δt: Inter-frame time interval; α and β are harmonic coefficients (α+β=1), used to balance the weights of velocity changes and energy changes.

[0066] The frame-skipping decision rules are as follows: Set a threshold Vth.

[0067] If V k If >Vth, then the k-th frame is determined to be a critical state frame and is retained.

[0068] If V kIf ≤Vth, then according to the preset maximum frame skipping interval N max Sampling is performed at fixed intervals (e.g., 1 frame is retained every N frames).

[0069] Figure 3 The diagram illustrates the steps of static point identification according to an embodiment of this application.

[0070] like Figure 3 As shown, S2 identifies static and dynamic sliding points within the landslide area, including: S21: Traverse the coordinates of points in consecutive keyframes and compare the 3D coordinates of points with the same index in two adjacent frames. S22: Determine it as a static point according to the determination rules, and save its index to the static point index list; S23: Based on the static point index list, assign different color and transparency attributes to static points and dynamic sliding points respectively during rendering.

[0071] The preferred determination rule for a point to be classified as a static point is as follows: if the maximum displacement of a point in M ​​consecutive keyframes is less than the preset inter-frame tolerance threshold ε, and the deviation between the centroid of its position sequence and its initial position is less than the overall drift tolerance threshold δ, then the point is determined to be a static point.

[0072] In preferred implementation, for example, the technical scenario involves automatically distinguishing between the stable, stationary parts (static points) and the displaced parts (dynamic points) within a landslide area by comparing consecutive keyframes.

[0073] The static point criterion based on spatiotemporal consistency is as follows: For data point j, its position sequence in M ​​consecutive keyframes is {P}. j 1, P j 2, ..., P j M}.

[0074] The conditions for determining a point to be static are as follows: max( || P j i - P j {i-1} || )<ε, for all i ∈ [2, M]; And || center of mass ({P) j 1, P j 2, ..., P j M}) - P j 1 ||<δ; The parameters are defined as follows: ||·|| represents the three-dimensional Euclidean distance; ε is the inter-frame displacement tolerance threshold, taking into account measurement noise or small numerical fluctuations; The centroid(...) is used to calculate the centroid of the position sequence.

[0075] Δ is the overall drift tolerance threshold, which requires that the point not only has small inter-frame changes, but also a stable overall position.

[0076] In practice, to improve the processing efficiency of massive point clouds, vectorized calculation is adopted to calculate the displacement matrix D (size: number of points × (number of frames - 1)) of all points in all adjacent frames at once. Then, by performing the matrix comparison operation D<ε, a static point Boolean index array is quickly generated.

[0077] Figure 4 The diagram illustrates the steps of constructing a three-dimensional geometric model according to an embodiment of this application.

[0078] like Figure 4 As shown, the steps for constructing a 3D geometric model in S3 include: S31: Convert geographic coordinates to Cartesian space coordinates and generate a vertex array; S32: Based on the static point index list and the dynamic sliding point set, assign preset color and transparency values ​​to the static and dynamic points in the vertex array, respectively. S33: Generate a triangle index array based on the sequence of vertex arrays to construct a landslide surface mesh for rendering; S34: Encapsulate the vertex array, color array, and index array into a Cesium.Geometry object, and create a corresponding GeometryInstance instance for rendering.

[0079] In a preferred embodiment, the method further includes: performing geographic location calibration on the landslide model, calculating the offset between the model's center point and the target geographic coordinates, and applying a spatial transformation matrix to achieve high-precision alignment between the model and the real terrain.

[0080] The dynamic visualization simulation in S4 also includes the following steps: creating a corresponding geometric instance for each keyframe and setting its display attributes; controlling the display and hiding of each geometric instance through a timer, playing them sequentially in time sequence to form a dynamic evolution effect of the landslide.

[0081] In step S4, "controlling the time sequence display and hiding of multiple geometric instances" is achieved through the Primitive object and GeometryInstance of the Cesium platform. Specifically, a GeometryInstance is created for the landslide surface state corresponding to each keyframe, and the ShowGeometryInstanceAttribute property of each GeometryInstance is dynamically switched through a timer to form a continuous animation playback effect.

[0082] Step S4 also includes a location calibration step: calculating the offset between the centroid of the landslide geometric model and the target geographic coordinates, and performing spatial transformation on the model by applying the Cesium.Matrix4 translation matrix to achieve accurate alignment with the real terrain.

[0083] Figure 5 , Figure 6 The diagram shows an overall flowchart of a three-dimensional dynamic visualization simulation method for landslide disasters according to an embodiment of this application.

[0084] like Figure 5 , Figure 6 As shown, the method of this application can be divided into four main stages: data source and cross-platform input, core data processing and conversion, 3D dynamic model construction and rendering, and application and output.

[0085] In Phase One, the application clearly distinguishes between "professional software data" and "monitoring equipment data," and in Phase Two, it is processed uniformly through a "cross-platform data adaptation engine," reflecting the core idea of ​​breaking down data silos.

[0086] Phase 1: Multi-source data and cross-platform compatibility.

[0087] First, data acquisition and preprocessing are performed. During the data acquisition phase, digital elevation model (DEM) data, geological structure data, and gridded terrain information of the mining area are systematically collected and integrated. Through spatial analysis, noise filtering, and interpolation algorithms, the terrain and geological elements are reconstructed with high precision. The preprocessed data is automatically converted into an elevation model format recognizable by the Cesium platform.

[0088] Based on this, the system performs standardized initialization configuration of dynamic parameters such as the initial location, volume, density, and friction coefficient of the landslide body, providing a unified physical basis and data benchmark for subsequent landslide dynamic calculations and three-dimensional dynamic visualization.

[0089] Regarding cross-software and cross-platform data conversion and adaptation mechanisms, in order to break through the data barriers between different software platforms in traditional landslide simulation, this application also proposes a cross-software and cross-platform data conversion and adaptation mechanism.

[0090] To address the data output formats of external professional mechanical simulation software, an automatic parsing, structural reconstruction, and format mapping process was designed. Through customized data bridging scripts, high-precision alignment of dynamic simulation results with the real geographical environment was achieved. This mechanism overcomes the data silo problem between different platforms, enabling simulation results to be reproduced in 3D and interactively analyzed against a realistic terrain background.

[0091] Phase Two: Core Data Processing and Transformation.

[0092] The original landslide observation data in this application is typically high-frequency time-series data, which is massive in volume, and the data frames at each time point do not contain geospatial coordinate information, only local relative coordinates. Directly loading this data into a 3D Geographic Information System (GIS) or visualization platform will lead to rendering delays, performance bottlenecks, and visualization discontinuities. To address this, this invention proposes a high-performance, automated spatial data generation and conversion method that can accurately map non-spatial coordinate data to actual geographic space, while optimizing the data structure to meet the real-time requirements of dynamic simulation and 3D rendering.

[0093] It mainly includes spatial coordinate transformation and high-performance preprocessing: 1. Spatial coordinate transformation methods and 2. High-frequency data frame extraction and downsampling strategies.

[0094] To address the problem that raw landslide observation data only contains local relative coordinates and lacks geospatial positioning, this invention proposes a high-performance spatial coordinate transformation method, including the following core processes: 1) Data centralization and spatial mapping; Local coordinate centering: The original data is uniformly placed in a local reference frame, with the centroid of the dataset as the origin, to construct a local geometric framework. This centering operation ensures that the relative relationships between data points are accurately preserved in the initial processing stage, while providing a unified benchmark for subsequent spatial mapping.

[0095] Key control point extraction and mapping correction: Several key geographic reference points (such as the landslide initiation point, endpoint, and intermediate control points) are selected using automatic or semi-automatic methods, and their positions in the original data coordinate system are matched with their corresponding actual geographic coordinates. Affine transformation, rigid body rotation, and nonlinear mapping algorithms are used to achieve high-precision mapping between the original coordinates and the geographic spatial coordinates; ensuring that non-spatial data accurately corresponds to the actual terrain, slope, and geographic location in the 3D GIS platform, thus providing a reliable spatial benchmark for subsequent landslide dynamics calculations and visualization simulations.

[0096] It also includes high-frequency data frame extraction and downsampling strategies as follows: The original observation system can generate about 60 frames of continuous data per second. If loaded in its entirety, it will severely impact rendering performance, increase memory usage, and reduce interactive response speed.

[0097] Dynamic frame extraction and downsampling: This application's embodiments design a scientific frame extraction method, selecting key frames for processing at fixed time steps, such as extracting 30 frames per second (i.e., extracting every other frame). This significantly reduces the amount of data while ensuring the continuity and dynamic smoothness of the landslide process. During frame extraction, the main movement trend and key change nodes of the landslide are preserved, making the simulation results both accurate and smooth. The frame extraction strategy can be flexibly adjusted according to the actual terrain complexity and simulation accuracy requirements, ensuring that the dynamic characteristics of landslides are fully reflected in different application scenarios.

[0098] Automated Scripting: To improve processing efficiency, this application employs a fully automated data conversion scripting system. Utilizing the high-performance Python programming language, it achieves fully automated processing from raw non-spatial coordinate data to spatialized GIS-loadable data. The script functions cover data reading, centralization, key point extraction, spatial mapping, frame extraction, format conversion, and output file generation. It supports batch processing of large-scale landslide event data and can automatically execute on datasets from different landslide areas and multiple time periods without manual intervention, significantly improving data processing efficiency and reliability.

[0099] Data Format Enhancement and Output: The converted data is in a standardized GIS-recognizable format (such as GeoJSON, Shapefile, or a custom 3D sequence data format), including the longitude, latitude, elevation, timestamp, and landslide feature information (color, volume, mass, etc.) for each data point. Through spatial coordinate appending and format standardization, the data can be directly loaded by the 3D visualization platform, supporting accurate rendering and high frame rate display of the landslide's dynamic evolution process.

[0100] Phase 3: Constructing a 3D dynamic model.

[0101] This application is built upon the core concept of "realistic geographic environment reconstruction + geographic coordinate-driven dynamic landslide simulation + 3D visualization rendering". The system includes a data loading and initialization module, a static point identification module, a geometric calculation and landslide modeling module, a dynamic rendering module, and a position calibration module. These modules work together to form a closed-loop process from raw observation data to dynamic simulation of realistic terrain.

[0102] First, data loading and initialization are performed; landslide keyframe data is read from the database or historical disaster data. Each keyframe contains a set of geographic coordinates (longitude, latitude, and elevation), describing the surface state of the landslide at a certain moment. The system passes this key data through the constructor and stores it in class properties for subsequent geometric calculations and animation control.

[0103] During initialization, a function is automatically called to compare data from different time frames in the landslide area. This process provides the basis for subsequent rendering of the sliding portion and differentiation of the static portion.

[0104] Implementation steps: (1) Load the keyframe data object through the constructor to automatically complete scene initialization and georeference alignment.

[0105] (2) Generate a list of static points based on the key frame sequence and call the static point recognition module to perform stable separation.

[0106] (3) Render a transparent initial landslide in a three-dimensional terrain (DEM) environment to realize the spatial occupancy of the landslide area and the binding of the terrain, providing a benchmark for subsequent dynamic evolution.

[0107] Then, static point identification and difference analysis are performed; the realism of the landslide simulation largely depends on the accurate division between the "sliding part" and the "stationary part". By comparing the coordinates of points in consecutive keyframes, it is determined which points remain unchanged in the time series, thereby generating a static point index list.

[0108] The core logic of the algorithm is to traverse all keyframe data, compare the coordinates of points with the same index in adjacent frames, and if the three-dimensional coordinates are completely identical, the point is determined to be a static point. The static point index is then saved for subsequent coloring and geometry generation. In this way, "stable regions" and "slipping regions" can be automatically distinguished in large-scale data, and thus differentiated in visualization using different transparency or colors, achieving an accurate representation of the landslide evolution range.

[0109] Implementation steps: 1) Traverse consecutive keyframe data and compare the changes in 3D coordinates point by point. 2) If the points in consecutive frames are consistent, they are identified as static points and their index positions are recorded. 3) Save the static point index sequence for subsequent sliding body geometry calculations and visualization difference rendering.

[0110] Next, geometric modeling and mesh generation of the landslide are carried out; after the static point identification is completed, geometric modeling is performed on the spatial point data of the landslide area.

[0111] The geometric modeling process includes three sub-processes: vertex calculation, color assignment, and index generation. Vertex calculation converts geographic coordinates (latitude, longitude, and elevation) into Cartesian 3 coordinates, forming a set of points in 3D space. Color assignment uses different color and transparency parameters for static and dynamic points; static points have lower transparency, while the color of sliding points dynamically adjusts according to elevation changes, creating a sense of sliding depth. Index generation traverses the point sequence, generating a triangular index array in row and column order to construct the mesh surface of the landslide area. This mesh not only represents the surface morphology of the landslide body but also enables dynamic deformation effects in subsequent rendering. This process is equivalent to constructing a deformable landslide model base mesh on the 3D terrain.

[0112] The implementation steps include: (1) Convert the latitude and longitude of each point in the observation data into geographic Cartesian coordinates to ensure consistency with the Earth reference system.

[0113] (2) Set geometric properties for sliding and static points respectively, including position, color, transparency and material type; the color of the sliding point changes with time to map slope and speed information.

[0114] (3) Generate a triangle index array through point sequence to form a continuous landslide surface grid and realize the physical combination of terrain and landslide body.

[0115] (4) Encapsulate the generated data into a Cesium.Geometry object, and create a GeometryInstance or Primitive instance for dynamic scene rendering.

[0116] Then, dynamic rendering and time-series control are performed; the key to landslide simulation lies in how to present the dynamic process of the landslide changing over time. The system uses Cesium.Primitive objects to manage multiple geometric instances (each instance corresponds to a keyframe state) and dynamically switches their visibility attribute (ShowGeometryInstanceAttribute) through a timer, thereby creating a "dynamic playback" effect of the landslide. The specific process is as follows: during initialization, multiple geometric instances are created (each instance represents the surface morphology of the landslide at different times), and each frame is displayed sequentially at time intervals (e.g., 1 second), with the previous frame hidden and the next frame displayed, achieving a continuous sliding effect.

[0117] The dynamic display of a landslide from initiation, flow to accumulation is achieved using time-series keyframes. The steps include: 1) Set display properties for each keyframe geometry instance.

[0118] 2) Start the timer to display the current frame and hide the previous frame at the set time interval.

[0119] 3) Finally, regarding location offset and interactive control: In order to make the landslide simulation model correspond to the real geographical location, after each rendering of geometry, the model is geographically calibrated by calculating the difference between the model's center point and the target's latitude and longitude.

[0120] To ensure accurate alignment between the model and real geographic coordinates, this module introduces a spatial matrix correction mechanism, the steps of which include: 1) Calculate the centroid location of the landslide model and compare it with the target geographic reference point.

[0121] 2) Generate a spatial offset matrix (Matrix4) based on the difference calculation to achieve accurate alignment between the sliding body model and the actual geographic coordinates.

[0122] 3) Adjust the model's posture and orientation in three-dimensional space in real time to make the landslide simulation results completely consistent with the actual terrain.

[0123] In terms of technical architecture, it makes full use of Cesium's WebGL rendering pipeline, Primitive geometry interface and Matrix4 matrix transformation capabilities to achieve lightweight and efficient dynamic simulation of landslides on the browser side.

[0124] Phase Four: Application and Output.

[0125] The high-fidelity, interactive, dynamic landslide digital twin generated through the first three stages of processing is transformed into a practical tool in the fourth stage that can directly serve mine safety management, disaster prevention and mitigation, and scientific research decision-making.

[0126] First, the three-dimensional dynamic visualization simulation method for landslide disasters, as described in this application, can generate a three-dimensional interactive dynamic landslide simulation scenario applicable to the web. This three-dimensional dynamic digital sand table can be accessed and operated in a standard web browser.

[0127] In the scene, the entire process of landslide occurrence, spread, and accumulation is presented in a continuous and smooth animation. The color and transparency of the landslide surface dynamically change with speed and position, clearly distinguishing the sliding zone from the stable zone. Users can freely rotate, zoom, and pan the viewpoint using the mouse and keyboard, and can pause, fast forward, or jump to any point in time for observation at any time.

[0128] This results in a complete transformation of the traditional analysis model based on two-dimensional drawings, static reports, or isolated videos. This application provides a common, intuitive, and in-depth spatiotemporal analysis foundation.

[0129] In terms of specific application functions: (1) Dynamic identification and impact analysis of risk areas; Real-time hazard delineation: During animation playback, the system can outline and display the leading edge boundary, maximum diffusion range, and affected area of ​​the landslide accumulation at different times. Managers can intuitively see "which road and facility will be affected by the landslide at X seconds."

[0130] Quantitative assessment of disaster intensity: Based on the attributes assigned to the landslide body during rendering (such as color representing velocity), combined with terrain data, areas with potentially high impact force and potential inundation / siltation depths can be automatically analyzed and displayed. For example, by analyzing the interaction between the landslide body and the river channel model, the potential reservoir capacity and dam height of the landslide-dammed lake can be quickly estimated.

[0131] Assisting in engineering planning: When planning new facilities or settlements, this system can be loaded to simulate the impact of potential landslide sources, providing an intuitive basis for risk avoidance in site selection and protection engineering design.

[0132] (2) Visual simulation and exercise of emergency response plans; Visualization of the plan: Established emergency plans (such as evacuation routes, safe havens, and storage points for rescue supplies) are overlaid onto a dynamic landslide scene as a 3D layer. Decision-makers can intuitively see whether the planned evacuation routes have been buried and whether the safe havens are safe at a certain stage of landslide development.

[0133] Dynamic simulation and optimization: By adjusting key landslide parameters (such as initiation area and volume), different disaster scenarios can be quickly simulated. Based on different simulation results, emergency command teams can dynamically simulate the feasibility and timeliness of plans such as personnel evacuation, traffic control, and engineering rescue, thereby optimizing existing plans and improving their scientific rigor and practicality.

[0134] Immersive training and drills: Without having to be physically present at the danger zone, relevant personnel can undergo immersive emergency response process training in a 3D scenario, become familiar with the disaster development process and key decision points, and improve their practical capabilities.

[0135] (3) Quantitative support for scientific research and engineering decision-making; Process inversion and mechanism analysis: Researchers can use this tool to perform high-precision inversion of historical landslide events. By adjusting parameters such as the friction coefficient and soil strength in the model, repeated simulations are conducted and compared with the actual accumulation range to analyze the key mechanical mechanisms and dominant factors of landslide occurrence in reverse.

[0136] Pre-simulation of mitigation effects: After designing anti-slide piles, retaining walls and other prevention and control projects, the engineering structure model can be imported into the system and the landslide simulation can be run again to simulate the stability of the mitigation project under landslide impact and its disaster reduction effect, thus realizing a closed loop of "design-simulation-optimization".

[0137] Data-driven decision-making reports: The system supports one-click export of scene screenshots at key time points, vector data of the impact range, impact energy distribution maps, etc., and automatically generates analysis reports containing key quantitative indicators (such as maximum velocity, total volume, and impact area), providing authoritative and intuitive data support for project review and risk assessment reports.

[0138] Therefore, through the embodiments of this application, a complete technical system was finally constructed, from multi-source data fusion to spatiotemporal process simulation and intelligent analysis and decision-making, realizing the ultimate goal of "visible, calculable, preventable and controllable" landslide disasters, and providing a core technical engine for smart mines and geological disaster prevention and control.

[0139] Figure 7 The diagram shows a dynamic simulation scenario of a landslide according to an embodiment of this application.

[0140] In practice, the process is as follows: Data Acquisition and Preprocessing Stage: The system first collects the digital elevation model (DEM), geological structure data, and gridded terrain information of the mining area, and then reconstructs the terrain and geological elements with high precision. The original high-frequency landslide observation data is usually in non-spatial local coordinates. To ensure it can be loaded into 3D GIS or Cesium platforms, the system centralizes the data, constructing a local reference frame with the dataset centroid as the origin. Simultaneously, by extracting key geographic control points and utilizing affine transformations, rigid body rotations, or nonlinear mappings, the system achieves high-precision mapping of the original data to actual geographic coordinates. For high-frequency time-series data (e.g., 60 frames per second), an intelligent frame extraction and downsampling strategy is adopted, retaining only key frames (e.g., 30 frames per second), significantly reducing data volume and rendering pressure while ensuring the dynamic continuity and physical characteristics of the landslide.

[0141] Furthermore, this phase emphasizes cross-platform and cross-software data adaptability: the system can parse the output results of external professional mechanical simulation software, perform structural reconstruction and format mapping, enabling data that could only run in specific numerical calculation software to be loaded, reproduced, and recalculated in a 3D visualization platform. This mechanism breaks down the barriers between dynamic simulation and visualization, achieving seamless integration between different software and data formats. The entire preprocessing process is fully automated through Python scripts, including data reading, spatial mapping, frame extraction, format conversion, and standardized output, generating standardized spatial data that can be directly loaded into GIS or 3D visualization platforms. This provides a high-precision, high-performance, repeatable, and cross-platform data foundation for landslide dynamics calculations and 3D rendering.

[0142] 3D Terrain Construction and Environment Loading: The system loads a terrain model (TerrainProvider) into the Cesium scene, uses terrain elevation to construct a realistic 3D mining area scene, and overlays auxiliary information such as mining area boundaries, roads, and drainage ditches through image layers and vector data loading functions to form a realistic geographical environment.

[0143] Landslide generation and parameter initialization: Calculate the initial volume of the landslide in the specified landslide area based on the terrain slope and soil thickness. Use polygons or point cloud particles to generate the landslide model in the Cesium scene and initialize the physical parameters of the landslide, such as gravitational acceleration, friction coefficient, initial velocity, time step Δt, etc.

[0144] Landslide Dynamic Calculation Module: This module uses a simplified motion equation based on physical laws to efficiently estimate the motion trend of the landslide body.

[0145] The sliding body can be generalized as a slider driven by gravity, whose motion along the slope follows the equation: F=m×a= m×g×sin(θ). μ×m×g×cos(θ); where m is the equivalent mass of the sliding body, a is the acceleration along the slope, g is the gravitational acceleration, θ is the local slope inclination angle queried in real time based on high-precision DEM data, and μ is the comprehensive friction coefficient.

[0146] The module continuously calculates the acceleration of the sliding body using this equation, then integrates it to obtain velocity and displacement, updating the sliding body's motion path on the 3D terrain. More importantly, this calculation process is tightly coupled with visualization; the calculated velocity values ​​are directly mapped to the color or transparency attribute of the sliding body model, achieving physical visualization of its motion state.

[0147] This simplified model is suitable for rapid simulation and trend analysis. For more complex dynamic processes, the system supports importing detailed simulation results from external professional mechanics software (such as PFC3D) via a cross-platform data interface for high-fidelity visualization and comparison, thus forming a full-scale simulation capability from rapid evaluation to detailed analysis.

[0148] The system calculates the acceleration, velocity, and displacement of the sliding body at each time step. The sliding body's motion path is updated by accumulating displacement, and the direction is adjusted in real time in combination with the terrain slope. The system updates the sliding body's position in each frame and determines whether it touches the ground, spreads out, or stops.

[0149] Visual rendering and dynamic performance: such as Figure 7 As shown, the dynamic process of the landslide is visualized using the WebGL rendering mechanism of Cesium. The trajectory of the landslide is represented by semi-transparent particles or gradient surfaces to indicate changes in speed and energy.

[0150] Example 2 This embodiment provides a three-dimensional dynamic visualization simulation system for landslide disasters. For details not disclosed in this embodiment, please refer to the specific implementation details of the three-dimensional dynamic visualization simulation schemes for landslide disasters in other embodiments.

[0151] Figure 8 The diagram shows a structural schematic of a three-dimensional dynamic visualization simulation system for landslide disasters according to an embodiment of this application.

[0152] like Figure 8 As shown, the three-dimensional dynamic visualization simulation system for landslide disasters includes: The data acquisition and preprocessing module 10 is used to acquire multi-source data of the landslide area, including digital elevation model (DEM) data, geological structure data, gridded topographic information, and physical parameters of the landslide body; The static point identification module 20 is used to identify static points and dynamic sliding points in the landslide area based on multi-source data, and generate a static point index list and a dynamic sliding point set. The landslide geometry modeling module 30 is used to construct a three-dimensional geometric model of the landslide area based on multi-source data, a static point index list, and a dynamic sliding point set, and to perform differentiated color and transparency rendering for static points and dynamic sliding points. The dynamic rendering and interactive control module 40 is used to load the three-dimensional geometric model into the Cesium-based three-dimensional visualization platform. Through time-series control and dynamic rendering, it realizes the dynamic visualization simulation of the entire process of landslide from initiation, flow to deposition in a real three-dimensional terrain environment.

[0153] Preferably, it also includes a location calibration module for aligning the landslide model with real geographic coordinates with high precision.

[0154] The three-dimensional dynamic visualization simulation system for landslide disasters proposed in this application acquires multi-source data of the landslide area, including digital elevation model (DEM) data, geological structure data, gridded terrain information, and physical parameters of the landslide body. Based on the multi-source data, static points and dynamic sliding points in the landslide area are identified. According to the multi-source data, the static point index list, and the dynamic sliding point set, a three-dimensional geometric model of the landslide area is constructed, and the static points and dynamic sliding points are rendered with differentiated colors and transparency. The three-dimensional geometric model is loaded into a Cesium-based three-dimensional visualization platform, and through time series control and dynamic rendering, the dynamic visualization simulation of the entire process of landslide from initiation, flow to deposition is realized in a real three-dimensional terrain environment.

[0155] This application integrates a 3D geographic information system, real-time dynamic simulation, and high-performance rendering to achieve fully automated processing, cross-platform data adaptation, and dynamic simulation in realistic geographic environments. This provides reliable and intuitive technical support for mine safety management, geological disaster prediction, and emergency decision-making. It solves key technical problems in visualization, accuracy, and performance of traditional landslide simulation.

[0156] Furthermore, this application utilizes the Web-based Cesium platform, employing technologies such as cross-platform data adaptation, high-frequency data processing, and differentiation between static points and dynamic sliding bodies. It also employs a frame-skipping downsampling strategy for high-frequency data, achieving the following technical effects: 1) Through a cross-platform adaptation mechanism, simulation data from different sources can be displayed in a unified Web 3D platform, lowering the barrier to entry and improving the system's versatility and ease of use. 2) Through a series of data preprocessing and rendering optimizations, large-scale, high-frequency landslide dynamic simulations can run smoothly in ordinary browsers, facilitating sharing and decision support; and high-performance web-based visualization has been achieved.

[0157] 3) Through high-precision spatial coordinate mapping and differentiated rendering of static / dynamic areas, the simulation results strictly correspond to the real terrain and can intuitively display the impact range of landslides, assisting in accurate analysis; thus enhancing the geospatial realism and analytical capabilities of the simulation results.

[0158] Example 3 This embodiment provides a three-dimensional dynamic visualization simulation device for landslide disasters. For details not disclosed in this embodiment, please refer to the specific implementation details of the three-dimensional dynamic visualization simulation method or system for landslide disasters in other embodiments.

[0159] Figure 9 The diagram shows a structural schematic of a three-dimensional dynamic visualization simulation device 400 for landslide disasters according to an embodiment of this application.

[0160] like Figure 9 As shown, the three-dimensional dynamic visualization simulation device 400 for landslide disasters includes: a storage unit 402 for storing executable instructions; and a processing unit 401 for connecting to the storage unit 402 to execute the executable instructions to complete the three-dimensional dynamic visualization simulation method for landslide disasters.

[0161] Those skilled in the art will understand that the illustration Figure 9 This is merely an example of a three-dimensional dynamic visualization simulation device 400 for landslide disasters and does not constitute a limitation on the three-dimensional dynamic visualization simulation device 400 for landslide disasters. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, the three-dimensional dynamic visualization simulation device 400 for landslide disasters may also include input / output devices, network access devices, buses, etc.

[0162] The processing unit 401 (Central Processing Unit, CPU) can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processing unit 401 can be any conventional processor. The processing unit 401 is the control center of the three-dimensional dynamic visualization simulation device 400 for landslide disasters, connecting all parts of the device via various interfaces and lines.

[0163] Storage unit 402 can be used to store computer-readable instructions. Processing unit 401 implements various functions of the three-dimensional dynamic visualization simulation device 400 for landslide disasters by running or executing the computer-readable instructions or modules stored in storage unit 402 and calling the data stored in storage unit 402. Storage unit 402 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the three-dimensional dynamic visualization simulation device 400 for landslide disasters, etc. In addition, storage unit 402 may include hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, read-only memory (ROM), random access memory (RAM), or other non-volatile / volatile storage devices.

[0164] If the integrated modules of the 3D dynamic visualization simulation device 400 for landslide disasters are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by instructing related hardware through computer-readable instructions. The computer-readable instructions can be stored in a computer-readable storage medium, and when executed by a processor, the computer-readable instructions can implement the steps of the various method embodiments described above.

[0165] Example 4 This embodiment provides a computer-readable storage medium on which a computer program is stored; the computer program is executed by a processor to implement a three-dimensional dynamic visualization simulation method for landslide disasters as described in other embodiments.

[0166] The three-dimensional dynamic visualization simulation device and storage medium for landslide disasters, as described in this application, includes acquiring multi-source data of the landslide area, including digital elevation model (DEM) data, geological structure data, gridded terrain information, and physical parameters of the landslide body; identifying static points and dynamic sliding points in the landslide area based on the multi-source data; constructing a three-dimensional geometric model of the landslide area based on the multi-source data, the static point index list, and the dynamic sliding point set, and rendering the static points and dynamic sliding points with differentiated colors and transparency; loading the three-dimensional geometric model into a Cesium-based three-dimensional visualization platform, and realizing dynamic visualization simulation of the entire process of landslide from initiation, flow to deposition in a real three-dimensional terrain environment through time series control and dynamic rendering.

[0167] This application integrates a 3D geographic information system, real-time dynamic simulation, and high-performance rendering to achieve fully automated processing, cross-platform data adaptation, and dynamic simulation in realistic geographic environments. This provides reliable and intuitive technical support for mine safety management, geological disaster prediction, and emergency decision-making. It solves key technical problems in visualization, accuracy, and performance of traditional landslide simulation.

[0168] Furthermore, this application utilizes the Web-based Cesium platform, employing technologies such as cross-platform data adaptation, high-frequency data processing, and differentiation between static points and dynamic sliding bodies. It also employs a frame-skipping downsampling strategy for high-frequency data, achieving the following technical effects: 1) Through a cross-platform adaptation mechanism, simulation data from different sources can be displayed in a unified Web 3D platform, lowering the barrier to entry and improving the system's versatility and ease of use. 2) Through a series of data preprocessing and rendering optimizations, large-scale, high-frequency landslide dynamic simulations can run smoothly in ordinary browsers, facilitating sharing and decision support; and high-performance web-based visualization has been achieved.

[0169] 3) Through high-precision spatial coordinate mapping and differentiated rendering of static / dynamic areas, the simulation results strictly correspond to the real terrain and can intuitively display the impact range of landslides, assisting in accurate analysis; thus enhancing the geospatial realism and analytical capabilities of the simulation results.

[0170] Those skilled in the art will understand that the terminology used in this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular forms “a,” “the,” and “the” as used in this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0171] It should be understood that although the terms first, second, third, etc., may be used in this invention to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of this invention, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0172] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0173] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A three-dimensional dynamic visualization simulation method for landslide disasters, characterized in that, Includes the following steps: Acquire multi-source data of the landslide area, including digital elevation model (DEM) data, geological structure data, gridded topographic information, and physical parameters of the landslide body; For the time series data of the original high-frequency landslide observation, key frames are adaptively extracted according to the rate of change of its motion state. By traversing the coordinates of points in consecutive key frames and comparing the three-dimensional coordinates of points with the same index in adjacent frames, static points and dynamic sliding points in the landslide area are identified according to the preset inter-frame displacement tolerance threshold and overall drift tolerance threshold, and a static point index list and a dynamic sliding point set are generated. Based on the multi-source data, the static point index list, and the dynamic sliding point set, a three-dimensional geometric model of the landslide area is constructed. Specifically, based on the static point index list and the dynamic sliding point set, preset and different color and transparency values ​​are directly assigned to the static points and dynamic sliding points in the constructed vertex array, so that the static area and the dynamic sliding area can be clearly distinguished visually. The three-dimensional geometric model is loaded into a Cesium-based three-dimensional visualization platform. Through time-series control and dynamic rendering, the entire process of landslide from initiation, flow to deposition is dynamically visualized and simulated in a real three-dimensional terrain environment.

2. The three-dimensional dynamic visualization simulation method for landslide disasters according to claim 1, characterized in that, The multi-source data also includes simulation output data from external mechanical simulation software; The method further includes: parsing, reconstructing and mapping the simulation output data into standardized data through a cross-platform data conversion and adaptation mechanism.

3. The three-dimensional dynamic visualization simulation method for landslide disasters according to claim 1, characterized in that, The method also includes: The multi-source data is preprocessed by converting the original non-spatial coordinate landslide observation data into standardized data with geographic spatial coordinates through data digitization, key control point mapping, and spatial coordinate transformation.

4. The three-dimensional dynamic visualization simulation method for landslide disasters according to claim 3, characterized in that, The process of converting raw, non-spatial coordinate landslide observation data into standardized data with geospatial coordinates through data digitization, key control point mapping, and spatial coordinate transformation includes: The landslide observation data is unified to a local reference frame with the centroid of the dataset as the origin; The starting point, ending point, and intermediate control point of the landslide were selected as key geographical reference points, and the correspondence between their local coordinates and actual geographical coordinates was established. Based on the aforementioned correspondence, the spatial transformation parameters are solved using the weighted least squares method, and all original data points are mapped to the real geographic coordinate system with high precision.

5. The three-dimensional dynamic visualization simulation method for landslide disasters according to claim 3, characterized in that, The preprocessing of the multi-source data further includes: High-frequency landslide observation data is subjected to frame extraction and downsampling processing. For the original high-frequency time-series data, key frames are adaptively extracted according to the rate of change of motion state. The rate of change of motion state is calculated based on the combined changes in the velocity of the center of mass of the landslide body and the changes in kinetic energy between adjacent frames. The entire process of data reading, coordinate mapping, frame extraction, and format conversion is automated through scripts.

6. The three-dimensional dynamic visualization simulation method for landslide disasters according to claim 4, characterized in that, The steps for constructing a three-dimensional geometric model of the landslide area include: Convert geographic coordinates to Cartesian space coordinates to generate a vertex array; Based on the static point index list and the dynamic sliding point set, preset color values ​​and transparency values ​​are assigned to the static points and dynamic points in the vertex array, respectively. A triangle index array is generated based on the sequence of the vertex array to construct a landslide surface mesh for rendering; The vertex array, color array, and index array are encapsulated into a Cesium.Geometry object, and a corresponding GeometryInstance instance is created for rendering.

7. A three-dimensional dynamic visualization simulation system for landslide disasters, characterized in that, include: The data acquisition and preprocessing module is used to acquire multi-source data of the landslide area, including digital elevation model (DEM) data, geological structure data, gridded topographic information, and physical parameters of the landslide body; The static point identification module is used to adaptively extract key frames from the time-series data of the original high-frequency landslide observation based on the rate of change of its motion state. By traversing the point coordinates in consecutive key frames and comparing the three-dimensional coordinates of points with the same index in adjacent frames, the module identifies static points and dynamic sliding points in the landslide area based on the preset inter-frame displacement tolerance threshold and overall drift tolerance threshold, and generates a static point index list and a dynamic sliding point set. The landslide geometric modeling module is used to construct a three-dimensional geometric model of the landslide area based on the multi-source data, the static point index list, and the dynamic sliding point set. Specifically, based on the static point index list and the dynamic sliding point set, preset different color and transparency values ​​are directly assigned to the static points and dynamic sliding points in the constructed vertex array, so that the static area and the dynamic sliding area can be clearly distinguished visually. The dynamic rendering and interactive control module is used to load the three-dimensional geometric model into the Cesium-based three-dimensional visualization platform. Through time-series control and dynamic rendering, it realizes the dynamic visualization simulation of the entire process of landslide from initiation, flow to deposition in a real three-dimensional terrain environment.

8. A three-dimensional dynamic visualization simulation device for landslide disasters, characterized in that, include: Storage unit, used to store executable instructions; And a processing unit, for connecting to a memory to execute executable instructions to complete the three-dimensional dynamic visualization simulation method for landslide disasters as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, It stores a computer program thereon; the computer program is executed by a processor to implement the method as described in any one of claims 1-6.

Images