An edge-computing-based real-time modeling method and system for field survey data

Abstract

The application provides a kind of based on edge computing's field survey data real-time modeling method and system, obtains the three-dimensional space point data of pre-set building group;Based on the surface curvature feature of three-dimensional space point data, three-dimensional space point data is dynamically spatial body unit division processing, to generate compressed spatial body unit data;Incremental modeling processing is carried out to compressed spatial body unit data, and structural discontinuous region is identified from incremental modeling processing result;The geometric topological relation of structural discontinuous region is repaired to space connection path, and continuous structure data is generated;Compressed spatial body unit data, incremental modeling processing result and continuous structure data are fused, and real-time updated building group three-dimensional model is generated.The application realizes the minute-level update of millimeter-level precision model in large building group field survey, breaks through the time efficiency bottleneck and structural continuity defect faced by traditional modeling method in limited equipment resource scene.

Description

Technical Field

[0001] This invention relates to the field of real-time modeling technology, and in particular to a method and system for real-time modeling of field survey data based on edge computing. Background Technology

[0002] As large-scale building complexes continue to expand in urban construction, and with increasing demands for efficiency in building safety monitoring, construction progress management, and emergency response, real-time modeling of on-site survey data has become a key technological aspect of the digital transformation of engineering projects. In continuous scanning scenarios, LiDAR or photogrammetry equipment continuously acquires 3D point cloud data of building complexes, necessitating high-precision, low-latency 3D model updates in dynamically changing environments to support core applications such as structural deformation analysis, space occupancy detection, and collaborative work planning.

[0003] Current solutions employ streaming modeling based on fixed voxel grids. This approach maps collected spatial point cloud data to a voxel grid of preset resolution, uses a hash table for sparse storage, and combines a sliding time window mechanism to reconstruct the local surface of newly added point clouds, achieving incremental model updates and alleviating computational pressure from data surges to some extent. However, existing solutions have some inherent drawbacks. These include the inability of fixed voxel granularity to adapt to differences in local geometric features, leading to modeling distortion or voids in structural transition regions or areas with dense detail; a lack of proactive identification and topological relationship repair mechanisms for spatial discontinuities, resulting in broken or misaligned models when the scanning perspective changes or occlusion occurs, affecting the coherent representation of the overall structure; and the data fusion stage relies solely on time-series overlay, failing to effectively integrate compressed data, incremental update results, and structural repair information, leading to lag and redundancy in model updates and making it difficult to meet the requirements of high-fidelity, real-time synchronous modeling. Summary of the Invention

[0004] This invention provides a real-time modeling method and system for field survey data based on edge computing. It addresses several issues in existing technologies, such as the inability of fixed voxel division granularity to adapt to differences in local geometric features, leading to modeling distortion or voids in structural transition areas or areas with dense details; the lack of proactive identification and topological relationship repair mechanisms for spatially discontinuous regions, resulting in broken or misaligned models when the scanning perspective changes or occlusion occurs, affecting the coherent representation of the overall structure; and the reliance on time-series overlay in the data fusion stage, failing to effectively integrate compressed data, incremental update results, and structural repair information, leading to lag and redundancy in model updates and making it difficult to meet the requirements of high-fidelity, real-time synchronous modeling.

[0005] In a first aspect, the present invention provides a real-time modeling method for field survey data based on edge computing, comprising:

[0006] Obtain 3D spatial point data of the pre-designed building complex;

[0007] Based on the surface curvature characteristics of the three-dimensional spatial point data, the three-dimensional spatial point data is dynamically divided into spatial volume units to generate compressed spatial volume unit data.

[0008] Incremental modeling is performed on the compressed spatial volume element data, and structural discontinuities are identified from the incremental modeling results.

[0009] Spatial connection path repair processing is performed on the geometric topological relationships of the discontinuous regions of the structure to generate continuous structural data;

[0010] The compressed spatial unit data, the incremental modeling results, and the continuous structure data are fused to generate a real-time updated 3D model of the building complex.

[0011] Optionally, obtain three-dimensional spatial point data of the pre-built building complex, including:

[0012] Collect motion trajectory data of a pre-set mobile scanning device in a pre-set building complex area, analyze the motion trajectory data, and generate a discrete trajectory point set;

[0013] The three-dimensional coordinate points of each building in the pre-set building complex are collected at the spatial locations corresponding to each discrete trajectory point in the discrete trajectory point set.

[0014] Establish the spatiotemporal correlation information between the discrete trajectory point set and the three-dimensional coordinate point set;

[0015] Based on the spatiotemporal correlation information, each three-dimensional coordinate point in the set of three-dimensional coordinate points is time-stamped to generate a sequence of three-dimensional coordinate points with timestamps.

[0016] The sequence of timestamped 3D coordinate points is divided into continuous scan segments, and multiple subsequences of 3D coordinate points with continuous scan paths are extracted from the processing results.

[0017] The subsequences of the three-dimensional coordinate points are merged to generate three-dimensional spatial point data.

[0018] Optionally, based on the surface curvature features of the three-dimensional spatial point data, dynamic spatial volume element partitioning is performed on the three-dimensional spatial point data to generate compressed spatial volume element data, including:

[0019] Calculate the surface curvature value within the neighborhood of each spatial point data item in the three-dimensional spatial point data;

[0020] The surface curvature values ​​are aggregated to generate a set of spatial point curvature values.

[0021] Based on the numerical distribution characteristics of the set of spatial point curvature values, the pre-set building complex is divided into a high curvature region and a low curvature region. Dense spatial meshing and sparse spatial meshing are then performed on the high curvature region and the low curvature region, respectively, to generate an initial set of spatial volume unit boundaries.

[0022] Based on the gradient change trend of the set of spatial point curvature values, the partition density transition parameter between adjacent spatial volume units in the initial set of spatial volume unit boundaries is adjusted to generate an optimized set of spatial volume unit boundaries.

[0023] Based on the optimized set of spatial volumetric unit boundaries, the three-dimensional spatial point data is subjected to spatial unit attribution mapping to generate a spatial point volumetric unit index relationship table.

[0024] Using the spatial point volume element index relationship table, calculate the geometric center coordinates of all spatial points within each spatial volume element, and generate a spatial volume element coordinate set as the compressed spatial volume element data.

[0025] Optionally, incremental modeling processing is performed on the compressed spatial volume element data, and structural discontinuities are identified from the incremental modeling results, including:

[0026] The newly added spatial volume unit set and the historical modeled spatial volume unit set are divided from the compressed spatial volume unit data;

[0027] Establish the spatial positional relationship between the newly added spatial volume unit set and the historical modeled spatial volume unit set to obtain a cross-period spatial adjacency table;

[0028] Calculate the surface curvature compatibility parameters of each pair of adjacent units in the cross-period spatial adjacency table, quantify each surface curvature compatibility parameter, and generate a set of inter-period compatibility metric values.

[0029] By applying the distribution characteristics of the compatibility metric set during the period, the adjacent unit pairs with compatibility below a preset compatibility threshold are identified as abnormal adjacent unit pairs. The abnormal adjacent unit pairs are aggregated to generate an initial set of broken unit pairs.

[0030] Spatial continuity analysis is performed on the initial fracture element set to generate structural fracture clusters;

[0031] Each of the structural fracture clusters is mapped onto the surface space of the pre-set building complex to generate structural discontinuity regions.

[0032] Optionally, spatial connection path repair processing is performed on the geometric topological relationships of the discontinuous regions of the structure to generate continuous structural data, including:

[0033] Identify the outer contour spatial volume elements of the discontinuous regions of the structure to generate a set of fracture boundary elements;

[0034] Using the spatial coordinates of each unit in the fracture boundary unit set, a topological connection skeleton is constructed as the initial connection path network;

[0035] Based on the preset structural mechanics constraints, a set of candidate repair paths is generated from the initial connection path network.

[0036] Calculate the spatial orientation compatibility parameters between the candidate repair path set and the set of building component units adjacent to the fracture boundary unit set, and aggregate the spatial orientation compatibility parameters to generate a path compatibility metric set.

[0037] Based on the set of path compatibility metrics, optimized repair paths that meet preset continuity conditions are selected from the set of candidate repair paths.

[0038] The optimized repair path is converted into a spatial triangular mesh structure, and the spatial triangular mesh structure is input into the discontinuous region of the structure to generate continuous structure data.

[0039] Optionally, spatial orientation compatibility parameters are calculated between the candidate repair path set and the set of building component units adjacent to the fracture boundary unit set. These spatial orientation compatibility parameters are then aggregated to generate a path compatibility metric set, including:

[0040] The spatial direction vectors of each candidate repair path in the candidate repair path set and the main direction vectors of each building component in the building component set adjacent to the fracture boundary unit set are projected and calculated to generate a path component projection relationship set.

[0041] The cosine of the angle between each of the spatial direction vectors and each of the principal direction vectors is calculated using the path component projection relationship set to generate a path direction compatibility value set.

[0042] Detect the spatial distance deviation between the endpoints of each candidate repair path and the geometric center point of the building component unit corresponding to each endpoint, and generate a path endpoint deviation set;

[0043] The set of path direction compatibility values ​​and the set of path endpoint deviations are weighted and fused to generate a single path compatibility parameter set.

[0044] The single path compatibility parameters are integrated based on their path identifiers in the single path compatibility parameter set to generate a set of path compatibility metrics.

[0045] Optionally, the compressed spatial unit data, the incremental modeling processing result, and the continuous structure data are fused to generate a real-time updated 3D model of the building complex, including:

[0046] The compressed spatial volume unit data is subjected to spatial occupancy mapping processing to generate a basic building geometric occupancy frame;

[0047] The structural change features in the incremental modeling process are input into the basic building geometric occupancy frame to generate an initial dynamically updated occupancy frame.

[0048] In the initial dynamic update placeholder framework, locate the spatial region corresponding to the continuous structural data, replace the spatial region with the spatial triangular mesh structure in the continuous structural data, and generate the target dynamic update placeholder framework.

[0049] The target dynamically updated placeholder framework is reconstructed using topology connections to generate a complete building topology network.

[0050] The spatial node coordinates of the complete building topology network are spatially triangulated to generate a set of triangular facets.

[0051] The triangular facet set is bound to the surface material features of the pre-set building complex to generate a real-time updated 3D model of the building complex.

[0052] Secondly, the present invention provides a real-time modeling system for field survey data based on edge computing, comprising:

[0053] The acquisition module is used to acquire three-dimensional spatial point data of a pre-set building complex;

[0054] The partitioning module is used to perform dynamic spatial volume unit partitioning on the three-dimensional spatial point data based on the surface curvature features of the three-dimensional spatial point data, so as to generate compressed spatial volume unit data.

[0055] The modeling module is used to perform incremental modeling processing on the compressed spatial volume unit data and identify structural discontinuities from the incremental modeling processing results.

[0056] The repair module is used to perform spatial connection path repair processing on the geometric topological relationships of the discontinuous regions of the structure to generate continuous structural data;

[0057] The fusion module is used to fuse the compressed spatial volume unit data, the incremental modeling processing results, and the continuous structure data to generate a real-time updated 3D model of the building complex.

[0058] Thirdly, the present invention provides a computing device including a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute a real-time modeling method for field survey data based on edge computing as described in any of the first aspects.

[0059] Fourthly, the present invention provides a computer storage medium storing computer program instructions thereon, wherein the computer program instructions, when executed by a processor, implement the real-time modeling method for field survey data based on edge computing as described in any one of the first aspects.

[0060] This invention acquires 3D point clouds of building complexes in real time using a mobile scanning device, and dynamically divides spatial volume units based on surface curvature at edge computing nodes to achieve data compression; identifies structural fracture areas based on the incremental compressed data, and generates topological repair paths in combination with building mechanics rules; finally, it integrates compressed data, incremental update results, and repaired structures to achieve minute-level updates of millimeter-precision models in on-site surveys of large building complexes, breaking through the timeliness bottleneck and structural continuity defects faced by traditional modeling methods in scenarios with limited equipment resources.

[0061] Furthermore, an initial path network is constructed by identifying structural fracture boundary elements, and candidate repair paths are derived based on beam-column load transfer rules. The orientation compatibility parameters between candidate paths and adjacent components are quantified, and optimized paths that meet mechanical continuity are selected and converted into triangular meshes. This solves the industry problem of conflict between geometric repair results and building structure physical rules, ensuring the topological continuity of the model while ensuring that the repair area meets building safety constraints such as load-bearing wall distribution and force flow transfer.

[0062] These or other aspects of the invention will become more apparent from the following description of the embodiments. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 A flowchart illustrating a real-time modeling method for field survey data based on edge computing, provided for an embodiment of the present invention;

[0065] Figure 2 A schematic diagram of the structure of a real-time modeling system for field survey data based on edge computing, provided in an embodiment of the present invention;

[0066] Figure 3 This is a schematic diagram of the structure of a computing device provided in an embodiment of the present invention. Detailed Implementation

[0067] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0068] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0069] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0070] Figure 1 A flowchart of a real-time modeling method for field survey data based on edge computing is provided in this embodiment of the invention, as shown below. Figure 1 As shown, the method includes:

[0071] The current field of building surveying faces three major technical challenges: First, the massive point cloud data collected by mobile scanning equipment exceeds the real-time processing limits of edge devices, and traditional uniform voxel compression damages surface features, leading to model distortion. Second, device pose drift during continuous scanning causes gaps in the point cloud, and general 3D reconstruction algorithms ignore the mechanical rules of building structures, generating fractured or load-bearing ineffective models. Third, centralized modeling solutions cannot achieve minute-level model updates in resource-constrained on-site environments, delaying construction decisions. To address these issues, this invention proposes the following approach: Through curvature-driven dynamic spatial unit partitioning technology, it achieves surface feature-preserving compression at edge nodes, overcoming the bottleneck of massive data processing; based on an incremental modeling mechanism, it captures structural fractures caused by scanning gaps, and combines building load transfer rules to generate topological repair paths, simultaneously addressing geometric continuity and mechanical compliance defects; finally, it integrates compressed data, incremental updates, and repaired structures to generate millimeter-level precision models, enabling end-to-end real-time modeling capabilities for large building complex on-site surveys, completely eliminating reliance on high-computing-power centers. Based on this, this invention provides a real-time modeling method for on-site survey data based on edge computing, such as... Figure 1 ,include:

[0072] Step 101: Obtain the three-dimensional spatial point data of the pre-set building complex.

[0073] In this step, the three-dimensional spatial point data refers to the set of three-dimensional coordinates of the building surface collected by the mobile scanning device, including position coordinates, scanning timestamps and device pose information flow, which is used to characterize the building geometry.

[0074] In this embodiment of the invention, the building complex area is first continuously scanned by a mobile scanning device to collect a set of three-dimensional coordinate points and corresponding motion trajectory data in real time; then, the motion trajectory data is parsed to generate a set of discrete trajectory points, and spatiotemporal correlation information between the discrete trajectory points and the corresponding three-dimensional coordinate point sets is established; then, a synchronization timestamp is marked for each three-dimensional coordinate point to form a timestamp sequence, and the scanning path is divided into a continuous set of point subsequences; finally, all subsequences are merged to generate three-dimensional spatial point data containing geometric features and pose information of the building surface.

[0075] Step 102: Based on the surface curvature features of the three-dimensional spatial point data, perform dynamic spatial volume unit division processing on the three-dimensional spatial point data to generate compressed spatial volume unit data.

[0076] In this step, surface curvature features refer to the quantified values ​​of surface curvature within the neighborhood of a spatial point, including a set of curvature values ​​obtained by calculating the local surface change rate of the point cloud, which is used to guide dynamic spatial partitioning; dynamic spatial volume element partitioning refers to the operation of dynamically adjusting the element partitioning density according to the curvature distribution, including the collaborative processing of dense partitioning in high curvature regions and sparse partitioning in low curvature regions, which is used to generate geometric feature-preserving compressed data; compressed spatial volume element data refers to the dimensionality-reduced dataset after dynamic partitioning, including the geometric center coordinates and element size parameters of each spatial volume element, which is used as input for incremental modeling.

[0077] In this embodiment of the invention, firstly, the curvature values ​​of the neighboring surfaces of each point in the three-dimensional spatial point data are calculated and aggregated to form a set of spatial point curvature values; secondly, high-curvature regions and low-curvature regions are divided according to the curvature value distribution, and dense and sparse spatial meshes are generated to form initial volume element boundaries respectively; then, the density transition parameters of adjacent volume elements are dynamically adjusted according to the curvature gradient to optimize the boundary set; finally, the three-dimensional points are mapped to the cells defined by the optimized boundaries, and the geometric center coordinates of the points within the cells are extracted to generate compressed spatial volume element data.

[0078] Step 103: Perform incremental modeling processing on the compressed spatial volume element data, and identify structural discontinuities from the incremental modeling processing results.

[0079] In this step, incremental modeling processing refers to a modeling mechanism that only processes newly added compressed data, including cross-cycle adjacency relationship construction, curvature compatibility analysis, and anomalous element clustering, used to identify structural fractures caused by scan gaps; incremental modeling processing results refer to the data set output by incremental analysis, including the spatial coordinate range of structural discontinuities and their associated anomalous element identifiers; structural discontinuities refer to fracture areas in the building model caused by scan gaps, including fracture boundary element sets and fracture spatial size parameters.

[0080] In this embodiment of the invention, firstly, the newly added unit set and the historical modeling unit set are distinguished from the current compressed spatial volume unit data; secondly, the spatial positional relationship between the newly added set and the historical set is established, and the cross-period adjacent unit pairs that meet the distance threshold are calculated to generate an adjacency relationship table; then, the surface curvature compatibility parameters of the adjacent unit pairs are calculated to form a set of metric values, and abnormal unit pairs with compatibility below the dynamic threshold are identified; finally, spatial continuity clustering analysis is performed on the abnormal unit pairs to map and generate structural discontinuities on the building surface.

[0081] Step 104: Perform spatial connection path repair processing on the geometric topological relationship of the discontinuous regions of the structure to generate continuous structural data.

[0082] In this step, geometric topology refers to the spatial connection rules of building components, including the force flow transmission direction of beam-column nodes and the planar continuity constraints of walls; spatial connection path repair processing refers to the operation of generating structural repair paths based on mechanical rules, including topological skeleton construction, candidate path derivation and orientation compatibility screening, used to reconstruct a continuous structure with effective load transfer; continuous structural data refers to the geometric model output by the repair processing, including the vertex coordinates, facet connection relationships and mechanical attribute labels of the triangular mesh structure.

[0083] In this embodiment of the invention, firstly, the outer contour spatial volume elements of the structural discontinuity region are identified to generate a set of fracture boundary elements; secondly, a topological connection skeleton is constructed based on the spatial coordinates of the boundary elements as an initial path network; subsequently, a set of candidate repair paths is derived according to the structural mechanics constraint rules, and the spatial orientation compatibility parameters between each path and adjacent component elements are calculated; then, the compatibility parameters are aggregated to generate a set of metrics, and optimized paths that meet the continuity conditions are selected. Finally, the optimized paths are converted into a triangular mesh structure and injected into the fracture region to generate continuous structural data.

[0084] Step 105: The compressed spatial unit data, the incremental modeling processing results, and the continuous structure data are fused to generate a real-time updated 3D model of the building complex.

[0085] In this step, the fusion operation refers to the process of integrating multi-source data, including updating the placeholder frame, replacing the repair area, and reconstructing the topology connection, which is used to generate a complete building model; the 3D model of the building complex refers to the final output visualization model, including a set of triangular facets, material texture mapping relationships, and a library of component mechanical parameters.

[0086] In this embodiment of the invention, firstly, spatial occupancy mapping is performed on the compressed spatial volume unit data to generate a basic geometric occupancy framework; secondly, structural change features from the incremental modeling results are injected into the basic framework to generate an initial dynamic update framework; then, the corresponding region of the continuous structural data is located, and the region is replaced with a triangular mesh structure to generate a target update framework; next, topological connection reconstruction is performed on the target framework to form a complete building topology network; finally, spatial triangulation is performed on the network node coordinates to generate a set of facets, and the building surface material features are bound to generate a real-time updated 3D model of the building complex.

[0087] For example, firstly, a backpack-style mobile scanning device continuously scans the commercial center building complex, collecting 3D coordinate point sets of columns, curtain walls, and roofs, as well as equipment movement trajectory data. The trajectories are analyzed to generate discrete trajectory point sets, and a spatiotemporal relationship is established with these point sets, forming timestamped 3D spatial point data. Secondly, the surface curvature values ​​of each region in the point data are calculated, dividing the curtain wall glass into high-curvature areas and the concrete wall into low-curvature areas. After dynamically dividing the spatial units, the geometric centers are extracted to generate compressed data. Then, the curvature compatibility of the newly added compressed units with the historical model is compared to identify structural fracture areas at the glass curtain wall joints. Next, a topological skeleton of the curtain wall boundary units is constructed, and candidate repair paths are generated according to the curtain wall keel bearing rules. Optimized paths compatible with the orientation of adjacent steel structures are selected and converted into triangular meshes. Finally, the column coordinates in the compressed data are mapped to placeholder frames, new elevator shaft structural features are injected, the curtain wall fracture areas are replaced with repair meshes, the topological connections are reconstructed, a triangular facet model is generated, and the stone material is bound to it. A real-time 3D model of the commercial center is output to guide the renovation construction.

[0088] This invention utilizes curvature-driven dynamic spatial unit partitioning technology to achieve high-fidelity compression of building point clouds on edge devices, overcoming the bottleneck of massive data processing. Based on an incremental modeling mechanism, it accurately locates structural fracture areas caused by scanning gaps, and generates effective repair paths for load transfer by combining building mechanics rules, eliminating the conflict between geometric continuity and structural safety. Finally, by fusing compressed data, incremental update results, and repaired structures, it generates real-time 3D models with millimeter-level precision in resource-constrained field environments, improving the surveying efficiency of large building complexes by more than 3 times and reducing the error reconstruction rate to 1%.

[0089] To address the spatiotemporal misalignment between trajectory data and point cloud acquisition during mobile scanning, this step generates high-precision 3D spatial point data through spatiotemporal correlation and continuous segment division. This invention provides a specific embodiment: Step 101, acquiring the 3D spatial point data of a pre-set building complex, specifically includes the following steps:

[0090] Step 111: Collect motion trajectory data of the preset mobile scanning device in the area of ​​the preset building complex, analyze the motion trajectory data, and generate a discrete trajectory point set.

[0091] In this step, motion trajectory data refers to the real-time position and attitude information flow of the mobile scanning device, including the device's three-dimensional coordinates, heading angle, pitch angle, and timestamp, which are used to reconstruct the scanning path; the parsing operation refers to the discretization processing of continuous trajectory data, including dividing the trajectory line at fixed time intervals, extracting discrete point coordinates, and adding time attributes; the discrete trajectory point set refers to the discrete spatial location set generated after parsing, with each point containing four-dimensional attributes: longitude, latitude, elevation, and timestamp.

[0092] In this embodiment of the invention, the inertial navigation system of the mobile scanning device first collects the motion trajectory data of the device in real time, which includes the device's position coordinates and attitude angle change sequence; secondly, the motion trajectory data is subjected to time domain discretization analysis to divide the continuous trajectory into discrete trajectory points with equal time intervals, and each discrete trajectory point contains three-dimensional coordinates and timestamp information; finally, a set of discrete trajectory points with spatial position and time attributes is generated.

[0093] Step 112: Collect the three-dimensional coordinate point set of each building in the preset building group at the spatial location corresponding to each discrete trajectory point in the discrete trajectory point set.

[0094] In this step, spatial location refers to the actual physical location of the device represented by discrete trajectory points, which is used to associate the point cloud data collected at that location; three-dimensional coordinate point set refers to the point cloud of the building surface collected at a specific device location, which includes the X coordinate value, Y coordinate value, Z coordinate value and reflection intensity information of multiple points.

[0095] In this embodiment of the invention, the spatial position of the device corresponding to each discrete trajectory point in the discrete trajectory point set is first determined; then, when the device is in each spatial position, the reflection point cloud of the building surface is collected by the laser scanning module to form a three-dimensional coordinate point set of the building surface corresponding to that position; finally, the position binding relationship between each three-dimensional coordinate point set and the discrete trajectory point is established.

[0096] Step 113: Establish the spatiotemporal correlation information between the discrete trajectory point set and the three-dimensional coordinate point set.

[0097] In this step, the spatiotemporal correlation information refers to the binding relationship between discrete trajectory points and a set of three-dimensional coordinate points, including the spatial position matching matrix and time synchronization correction parameters.

[0098] In this embodiment of the invention, firstly, a mapping relationship is established between the time series of discrete trajectory points and the acquisition time of the three-dimensional coordinate point set; secondly, the data transmission delay is compensated according to the device movement speed, and each three-dimensional coordinate point set is associated with the nearest neighbor discrete trajectory point; finally, spatiotemporal association information containing position correspondence and time synchronization relationship is generated.

[0099] Step 114: Based on the spatiotemporal correlation information, perform synchronous timestamp marking on each three-dimensional coordinate point in the three-dimensional coordinate point set to generate a sequence of three-dimensional coordinate point sets with timestamps.

[0100] In this step, the synchronization timestamp marking process refers to the operation of injecting the timestamps of discrete trajectory points into the corresponding three-dimensional coordinate point set, so that each point carries the acquisition time information; the sequence of three-dimensional coordinate point sets with timestamps refers to the set of point sets arranged in chronological order, and all points in each point set share the same acquisition timestamp.

[0101] In this embodiment of the invention, firstly, the timestamps of discrete trajectory points corresponding to each set of three-dimensional coordinate points are determined based on spatiotemporal correlation information; secondly, the timestamps are marked in the data record of each three-dimensional coordinate point in the corresponding set; finally, a sequence of three-dimensional coordinate point sets with timestamps arranged in chronological order is formed.

[0102] Step 115: Perform continuous scan segment division processing on the timestamped three-dimensional coordinate point set sequence, and extract multiple scan path continuous three-dimensional coordinate point set subsequences from the processing result.

[0103] In this step, the continuous scan segment division process refers to the operation of dividing the point set sequence according to the device's start and stop status, including detecting scan interruption signal points and delineating the boundaries of continuous segments; the processing result refers to the intermediate dataset generated after division, which contains the start and end times of multiple independent continuous scan segments and the associated point set identifiers; the scan path refers to the spatial trajectory line formed by the uninterrupted movement of the device, and the corresponding point set has continuous positional connection characteristics; the three-dimensional coordinate point set subsequence refers to the point set sequence within a single continuous scan segment, where the device's motion trajectory within the segment is continuous and there are no pose jumps.

[0104] In this embodiment of the invention, the device start-stop signal points in the sequence of timestamped three-dimensional coordinate points are first detected; then, the sequence is divided into multiple continuous scanning segments using the start-stop points as the dividing boundary; subsequently, the three-dimensional coordinate point set subsequences within each continuous scanning segment are extracted to ensure that the device motion trajectory within the subsequences is continuous and uninterrupted; finally, a set of three-dimensional coordinate point set subsequences with continuous scanning paths is output.

[0105] Step 116: Merge the subsequences of the three-dimensional coordinate points to generate three-dimensional spatial point data.

[0106] In this step, the merging operation refers to the integrated processing of multiple subsequences, including time axis alignment, point cloud registration of overlapping regions, and global coordinate system unification.

[0107] In this embodiment of the invention, the subsequences of each three-dimensional coordinate point set are first arranged in time stamp order; then, spatial position registration is performed on the overlapping areas of adjacent subsequences to eliminate device pose drift error; finally, the point cloud data of all subsequences are merged to generate complete three-dimensional spatial point data.

[0108] This invention addresses the spatial misalignment problem caused by pose drift during mobile scanning by precisely correlated motion trajectories with point cloud data in a spatiotemporal manner; it eliminates interference from invalid data during scanning intervals by using a continuous scanning segment division mechanism based on the device's start and stop states; and it generates high-precision three-dimensional spatial point data by registering and merging overlapping areas, reducing the error in building surface reconstruction to the millimeter level, thus providing a complete geometric feature input source for dynamic spatial unit division.

[0109] To address the problem of uniform voxel compression damaging the surface features of buildings, this step achieves feature-preserving compression by dynamically dividing spatial volume elements based on curvature distribution. A specific embodiment of this invention is provided: Step 102, based on the surface curvature features of the three-dimensional spatial point data, performs dynamic spatial volume element division processing on the three-dimensional spatial point data to generate compressed spatial volume element data, specifically including the following steps:

[0110] Step 201: Calculate the surface curvature value within the neighborhood of each spatial point data item in the three-dimensional spatial point data.

[0111] In this step, the spatial point data item refers to a single data record in the three-dimensional spatial point data, which includes the three-dimensional coordinates and attribute information of the point; the neighborhood range refers to the spatial region within a fixed radius centered on the target point, which is used to calculate local geometric features; the surface curvature value refers to the curvature quantization value calculated by the normal change rate of the neighborhood points, which reflects the local surface unevenness characteristics.

[0112] In this embodiment of the invention, firstly, each spatial point data item in the three-dimensional spatial point data is selected, and the spatial coordinates of all neighboring points within its neighborhood are determined; secondly, the surface curvature value is quantified by calculating the rate of change of the local surface normal formed by the point and its neighboring points; finally, the surface curvature value corresponding to each spatial point is output.

[0113] Step 202: Aggregate the surface curvature values ​​to generate a set of spatial point curvature values.

[0114] In this step, the aggregation operation refers to the normalization process of discrete curvature values, including outlier filtering and data format standardization; the spatial point curvature value set refers to a data structure containing all point curvature values, organized in spatial order.

[0115] In this embodiment of the invention, firstly, the surface curvature values ​​of all spatial points are collected; secondly, data aggregation is performed in spatial order to eliminate abnormal fluctuation values; finally, a set of spatial point curvature values ​​containing the curvature characteristics of each location on the building surface is generated.

[0116] Step 203: Based on the numerical distribution characteristics of the set of spatial point curvature values, the pre-set building complex is divided into a high curvature region and a low curvature region. Dense spatial meshing and sparse spatial meshing are performed on the high curvature region and the low curvature region respectively to generate an initial set of spatial volume unit boundaries.

[0117] In this step, numerical distribution characteristics refer to the statistical properties of curvature values ​​on the building surface, including maximum, minimum, and concentrated distribution intervals; high curvature regions refer to surface areas where curvature values ​​exceed a set threshold, such as the edges of architectural decorative components; low curvature regions refer to surface areas where curvature values ​​are below a set threshold, such as the plane of a building wall; dense spatial mesh generation refers to the operation of generating small-sized spatial units in high curvature regions; sparse spatial mesh generation refers to the operation of generating large-sized spatial units in low curvature regions; and the initial spatial unit boundary set refers to the set of three-dimensional boundary coordinates of the initially divided spatial units.

[0118] In this embodiment of the invention, the numerical distribution characteristics of the set of curvature values ​​of spatial points are first analyzed, and a curvature threshold is set to divide the high curvature region into a low curvature region. Secondly, a fine spatial mesh is performed on the high curvature region, and a coarse spatial mesh is performed on the low curvature region. Finally, an initial set of spatial volume element boundaries is generated, wherein the element size in the high curvature region is small, and the element size in the low curvature region is large.

[0119] Step 204: Based on the gradient change trend of the set of spatial point curvature values, adjust the partition density transition parameters between adjacent spatial units in the initial set of spatial unit boundaries to generate an optimized set of spatial unit boundaries.

[0120] In this step, the gradient change trend refers to the direction and rate of curvature value change in adjacent regions; the spatial volume unit refers to the smallest processing unit in the three-dimensional space formed by partitioning; the partition density transition parameter refers to the adjustment coefficient that controls the rate of change of adjacent unit size; the adjustment operation refers to the process of optimizing the unit boundary based on the curvature gradient; and the optimized spatial volume unit boundary set refers to the unit boundary dataset that has been adaptively adjusted by curvature.

[0121] In this embodiment of the invention, the gradient change trend of the set of spatial point curvature values ​​in adjacent regions is first detected; then, the partition density transition parameters of adjacent units in the initial spatial volume unit boundary set are adjusted according to the curvature gradient direction to make the unit size transition smoothly; finally, the optimized spatial volume unit boundary set with adaptive curvature change is output.

[0122] Step 205: Perform spatial unit attribution mapping on the three-dimensional spatial point data according to the optimized spatial volume unit boundary set, and generate a spatial point volume unit index relationship table.

[0123] In this step, the spatial unit attribution mapping process refers to the operation of associating point coordinates with their respective spatial volume units; the spatial point volume unit index relationship table refers to the data structure that records the attribution relationship between points and units.

[0124] In this embodiment of the invention, firstly, the three-dimensional boundary range of each spatial volume unit is determined based on the optimized spatial volume unit boundary set; secondly, each spatial point in the three-dimensional spatial point data is mapped to its corresponding spatial volume unit; finally, a spatial point volume unit index relationship table is generated to record the relationship between the points and the units.

[0125] Step 206: Apply the spatial point volume element index relationship table to calculate the geometric center coordinates of all spatial points within each spatial volume element, and generate a spatial volume element coordinate set as the compressed spatial volume element data.

[0126] In this step, a spatial point refers to the original three-dimensional coordinate point on the building surface; the geometric center coordinate refers to the average coordinate of all points within a spatial unit; and the spatial unit coordinate set refers to the set of geometric center coordinates of all units.

[0127] In this embodiment of the invention, firstly, all spatial points within each spatial volume unit are retrieved through the spatial point volume unit index relationship table; secondly, the geometric center coordinates of the spatial points within the unit are calculated; finally, the geometric center coordinates of all units are aggregated to generate a spatial volume unit coordinate set, which is output as compressed spatial volume unit data.

[0128] This invention employs curvature-driven dynamic spatial unit partitioning technology to compress data while preserving architectural and decorative details; it optimizes the transition region of units based on curvature gradient to eliminate feature edge distortion caused by traditional uniform partitioning; and finally generates geometrically feature-preserving compressed data, thereby improving the efficiency of large building point cloud processing by five times and reducing feature reconstruction error to the millimeter level.

[0129] To address the issue of inaccurate identification of seam breaks between old and new models caused by scanning gaps, this step locates structural discontinuities through cross-period curvature compatibility analysis. This invention provides a specific embodiment: Step 103 involves incremental modeling of the compressed spatial volume element data, identifying structural discontinuities from the incremental modeling results, specifically including the following steps:

[0130] Step 301: Divide the newly added spatial volume unit set and the historical modeling spatial volume unit set from the compressed spatial volume unit data.

[0131] In this step, the newly added spatial unit set refers to the compressed spatial unit dataset generated in the current scanning cycle that has not participated in the historical modeling; the historical modeling spatial unit set refers to the compressed spatial unit historical dataset that has been integrated into the architectural 3D model.

[0132] In this embodiment of the invention, firstly, a subset of newly added units that did not participate in historical modeling is identified from the current compressed spatial unit data; secondly, a subset of historical modeling units that has been integrated into the building model is extracted; and finally, a set of newly added spatial units and a set of historical modeling spatial units are generated.

[0133] Step 302: Establish the spatial positional relationship between the newly added spatial volume unit set and the historical modeling spatial volume unit set to obtain a cross-period spatial adjacency table.

[0134] In this step, spatial position relationship refers to the three-dimensional spatial distance matrix and orientation angle information between the newly added and historical units; cross-period spatial adjacency relationship table refers to the data structure that records the information of adjacent unit pairs collected in different time periods.

[0135] In this embodiment of the invention, the three-dimensional spatial distance between each unit in the newly added spatial volume unit set and the historical modeled spatial volume unit set is first calculated; then, cross-period unit pairs that meet the preset adjacency distance threshold are selected; finally, spatial positional relationships are established to generate a cross-period spatial adjacency relationship table.

[0136] Step 303: Calculate the surface curvature compatibility parameters of each adjacent unit pair in the cross-period spatial adjacency relationship table, quantify each surface curvature compatibility parameter, and generate a set of inter-period compatibility metric values.

[0137] In this step, adjacent cell pairs refer to pairs of cells whose spatial distance is less than a preset threshold, including one newly added cell and one historical cell; the surface curvature compatibility parameter refers to the quantitative index of the difference in surface curvature values ​​between adjacent cells, reflecting geometric continuity; quantization refers to the operation of normalizing the compatibility parameter to a standard value in the range of zero to one; the periodic compatibility measurement value set refers to the set of normalized compatibility values ​​of all adjacent cells.

[0138] In this embodiment of the invention, each pair of adjacent cells in the cross-period spatial adjacency table is first obtained; then, the surface curvature compatibility parameters between newly added cells and historical cells in the cell pair are calculated; subsequently, the parameters are normalized and quantized; finally, a set of period compatibility metric values ​​containing the compatibility values ​​of all cell pairs is generated.

[0139] Step 304: Apply the distribution characteristics of the compatibility metric set during the period to identify adjacent unit pairs with compatibility below a preset compatibility threshold as abnormal adjacent unit pairs, aggregate the abnormal adjacent unit pairs to generate an initial set of broken unit pairs.

[0140] In this step, distribution characteristics refer to the statistical properties of the compatible value set, including mean, variance, and concentration range; compatibility refers to the degree of geometric feature matching between new and old units at the connection point; the preset compatibility threshold refers to the dynamic threshold value for determining whether a unit pair is broken, which varies with the data distribution; abnormal adjacent unit pairs refer to unit pairs with compatibility below the threshold, representing potential break locations; aggregation operation refers to the operation of merging spatially adjacent abnormal unit pairs; the initial broken unit pair set refers to the spatially associated abnormal unit group formed after aggregation.

[0141] In this embodiment of the invention, the numerical distribution characteristics of the compatibility metric set during the period are first analyzed; then, a dynamic compatibility threshold is set to identify abnormal adjacent unit pairs below the threshold; subsequently, a spatial proximity aggregation operation is performed on the abnormal unit pairs; finally, an initial set of broken unit pairs with spatial location association is generated.

[0142] Step 305: Perform spatial continuity analysis on the initial fracture element set to generate structural fracture clusters.

[0143] In this step, spatial continuity analysis refers to the processing of detecting the spatial distribution continuity of anomalous unit groups; structural fracture clusters refer to the set of fracture regions composed of spatially continuous anomalous unit groups.

[0144] In this embodiment of the invention, spatial clustering analysis is first performed on the initial set of fracture unit pairs; then, unit pairs with a spatial distance less than the fracture threshold are merged; finally, structural fracture clusters representing continuous fracture regions are generated.

[0145] Step 306: Map each of the structural fracture clusters to the surface space of the pre-set building complex to generate structural discontinuity regions.

[0146] In this step, the mapping operation refers to the process of transforming the coordinates of the fracture clusters to the coordinate system of the building surface; the surface space refers to the three-dimensional curved space formed by the building facade and roof.

[0147] In this embodiment of the invention, the spatial coordinate range of the structural fracture cluster is first extracted; then the coordinate range is mapped to the surface space of the building complex; finally, a structural discontinuity region containing location boundaries and fracture degree parameters is generated.

[0148] This invention accurately locates structural fractures between new and historical models through cross-period adjacency relationship analysis; it eliminates misjudgments caused by device pose drift based on a curvature compatibility dynamic threshold recognition mechanism; and finally generates spatially continuous building structural fracture areas, providing millimeter-level precision input for geometric repair, thus significantly improving the model update efficiency of large-scale building renovation projects.

[0149] To address the issue of geometric repair paths violating structural mechanics rules, this step combines orientation compatibility quantification to generate a continuous structure with effective load transfer. This invention provides a specific embodiment: Step 104 involves performing spatial connection path repair processing on the geometric topological relationships of the discontinuous structural regions to generate continuous structural data, specifically including the following steps:

[0150] Step 401: Identify the outer contour spatial volume elements of the discontinuous regions of the structure to generate a set of fracture boundary elements.

[0151] In this step, the outer contour spatial volume unit refers to the outermost spatial volume unit of the structural discontinuity region, which includes the boundary geometric features of the fracture region; the fracture boundary unit set refers to the set composed of outer contour units, which characterizes the spatial extent of the fracture region.

[0152] In this embodiment of the invention, firstly, edge space volume elements of the structural discontinuity region are detected, and all elements constituting the outer contour of the fracture region are identified; secondly, elements with significant curvature changes in the outer contour elements are screened, and internal non-boundary elements are excluded; finally, a complete set of fracture boundary elements surrounding the fracture region is generated.

[0153] Step 402: Using the spatial coordinates of each unit in the fracture boundary unit set, construct a topological connection skeleton as the initial connection path network.

[0154] In this step, the topological connection skeleton refers to the wireframe network formed by connecting the geometric centers of the fracture boundary units, reflecting the connectivity of the fractured region; the initial connection path network refers to the set of repairable paths formed by the transformation of the topological skeleton, which includes basic connection segments.

[0155] In this embodiment of the invention, the three-dimensional geometric center coordinates of each unit in the fracture boundary unit set are first obtained; then, the geometric centers of adjacent units are connected to form a spatial line segment network; finally, a topological connection skeleton representing the connectivity of the fracture region is constructed as an initial connection path network.

[0156] Step 403: Generate a set of candidate repair paths based on the initial connection path network according to the preset building structure mechanics constraint rules.

[0157] In this step, the preset building structural mechanics constraint rules refer to the building component load-bearing rule library, including minimum load-bearing angle and load transfer direction constraints; the candidate repair path set refers to the basic repair path dataset that conforms to the mechanics rules.

[0158] In this embodiment of the invention, a preset building structure mechanics constraint rule library is first loaded, including beam and column load-bearing rules and shear force transfer rules; secondly, candidate paths that meet the minimum radius of curvature are generated in the initial connection path network; finally, paths that violate the load transfer direction are excluded to form a set of candidate repair paths.

[0159] Step 404: Calculate the spatial orientation compatibility parameters between the candidate repair path set and the building component unit set adjacent to the fracture boundary unit set, and aggregate the spatial orientation compatibility parameters to generate a path compatibility metric set.

[0160] In this step, the building component unit set refers to the set of beam, column and wall units adjacent to the fracture area, which includes spatial orientation attributes; the spatial orientation compatibility parameter refers to the quantified value of the consistency between the repair path and the direction of adjacent components; the aggregation operation refers to the weighted summation calculation of multiple compatibility parameters; and the path compatibility metric set refers to the standardized dataset of all path compatibility parameters.

[0161] In this embodiment of the invention, firstly, the set of building component units adjacent to the fracture boundary unit set is retrieved; secondly, the cosine value of the angle between the candidate repair path direction vector and the main direction vector of the component is calculated; then, the distance deviation between the path endpoint and the geometric center of the component is detected; finally, the angle value and the distance value are weighted and fused to generate spatial orientation compatibility parameters, and all parameters are aggregated to form a set of path compatibility measurement values.

[0162] Step 405: Based on the set of path compatibility metrics, select optimized repair paths that meet preset continuity conditions from the set of candidate repair paths.

[0163] In this step, the preset continuity condition refers to the threshold of dual requirements for mechanical compliance and geometric smoothness; the optimized repair path refers to the final repair path that simultaneously satisfies mechanical rules and geometric compatibility.

[0164] In this embodiment of the invention, the numerical distribution range of the path compatibility metric set is first analyzed; then, a continuity condition threshold is set to filter paths that meet the compatibility criteria; finally, optimized repair paths that satisfy both mechanical and geometric continuity are extracted from the candidate repair path set.

[0165] Step 406: Convert the optimized repair path into a spatial triangular mesh structure, input the spatial triangular mesh structure into the discontinuous region of the structure, and generate continuous structure data.

[0166] In this step, the transformation operation refers to the process of converting a linear path into a surface mesh; the spatial triangular mesh structure refers to a surface geometric model composed of triangular facets, including vertex coordinates and facet connection relationships.

[0167] In this embodiment of the invention, the optimized repair path is first discretized into a series of spatial node coordinates; then, the nodes are connected to form a triangular patch network; finally, the spatial triangular mesh structure is covered to the discontinuous area of ​​the structure, replacing the original fracture unit to generate continuous structural data.

[0168] This invention generates an initial repair path through a fracture boundary topological skeleton and selects candidate paths by combining them with building mechanics rules. It innovatively introduces a spatial orientation compatibility quantification mechanism to simultaneously ensure the geometric continuity and structural safety of the repair path. Finally, it generates a triangular mesh repair structure with effective load transfer, solving the industry problem of conflict between aesthetic repair and load-bearing requirements in traditional methods.

[0169] To address the lack of quantitative data on the compatibility between repair paths and adjacent components, this step generates standardized compatibility metrics through vector projection and deviation fusion. A specific embodiment of this invention is provided: Step 404, calculating the spatial orientation compatibility parameters between the candidate repair path set and the set of adjacent building component units of the fracture boundary unit set, and aggregating the spatial orientation compatibility parameters to generate a path compatibility metric set, specifically including the following steps:

[0170] Step 441: Perform projection calculation on the spatial direction vector of each candidate repair path in the candidate repair path set and the main direction vector of each building component in the building component set adjacent to the fracture boundary unit set to generate a path component projection relationship set.

[0171] In this step, the spatial direction vector refers to the extended direction vector of the candidate repair path, which includes the direction components in three-dimensional space; the principal direction vector refers to the dominant direction vector of the load-bearing structure of the building component unit, which is determined by the geometric characteristics of the component; the projection calculation process refers to the operation of calculating the dot product of two vectors, which reflects the degree of consistency between the directions of the vectors; the path component projection relationship set refers to the data set that stores the dot product values ​​of the path vector and the component vector.

[0172] In this embodiment of the invention, the spatial direction vectors of each path in the candidate repair path set and the principal direction vectors of adjacent building component units are first obtained. Next, a dot product projection calculation is performed on each pair of path vectors and component vectors. Finally, a path-component projection relationship set recording the projection values ​​of all vector pairs is generated.

[0173] Step 442: Calculate the cosine of the angle between each of the spatial direction vectors and each of the principal direction vectors using the path component projection relationship set to generate a path direction compatibility value set.

[0174] In this step, the cosine value of the included angle refers to the cosine function value of the angle between two vectors, which is calculated by dividing the dot product by the product of the magnitudes; the path direction compatibility value set refers to the set of quantized values ​​of the compatibility between all paths and component directions.

[0175] In this embodiment of the invention, firstly, the dot product value and vector magnitude of each vector pair are extracted from the path component projection relationship set. Secondly, the ratio of the dot product value to the product of the two vector magnitudes is calculated to obtain the cosine value of the included angle. Finally, all cosine values ​​are aggregated to generate a path direction compatibility value set.

[0176] Step 443: Detect the spatial distance deviation between the endpoints of each candidate repair path and the geometric center point of the building component unit corresponding to each endpoint, and generate a path endpoint deviation set.

[0177] In this step, endpoints refer to the coordinates of the start and end points of the candidate repair path; geometric center point refers to the average coordinate position of all points within the building component unit; spatial distance deviation value refers to the straight-line distance between the path endpoint and the component center point; and path endpoint deviation set refers to the set of all path endpoint distance deviation values.

[0178] In this embodiment of the invention, the coordinates of the endpoints of the candidate repair path and the coordinates of the geometric center point of their corresponding building component unit are first located. Next, the Euclidean spatial distance deviation between the endpoints and the geometric center point is calculated. Finally, a path endpoint deviation set containing all path deviation values ​​is generated.

[0179] Step 444: Perform weighted fusion processing on the path direction compatibility value set and the path endpoint deviation set to generate a single path compatibility parameter set.

[0180] In this step, weighted fusion processing refers to the process of comprehensively calculating the direction compatibility value and the distance deviation value according to the weight ratio; the single path compatibility parameter set refers to the comprehensive compatibility parameter dataset calculated independently for each path.

[0181] In this embodiment of the invention, firstly, directional weight coefficients are assigned to the set of path direction compatibility values. Secondly, distance weight coefficients are assigned to the set of path endpoint deviations. Subsequently, a weighted summation calculation is performed on the compatibility value and deviation value for each path. Finally, a single-path compatibility parameter set is output.

[0182] Step 445: Integrate each single path compatibility parameter according to the path identifier of each single path compatibility parameter in the single path compatibility parameter set to generate a path compatibility metric set.

[0183] In this step, the path identifier refers to a unique numbered label that identifies a candidate repair path; the integration operation refers to the process of grouping and summarizing compatibility parameters by path identifier.

[0184] In this embodiment of the invention, firstly, the path identifiers of each parameter in the single-path compatibility parameter set are extracted. Secondly, the compatibility parameters are grouped and integrated according to their identifiers. Finally, a set of path compatibility metrics is generated, with each path identifier mapped one-to-one with a compatibility parameter.

[0185] This invention uses vector projection and cosine angle calculation to accurately quantify the directional compatibility between paths and components; combined with a weighted fusion mechanism of endpoint distance deviation, it generates a dual compatibility index that takes into account both geometric orientation and spatial location; ultimately forming a standardized set of path compatibility metrics, providing a quantifiable decision-making basis for the safe repair of building structures.

[0186] To address the issue of topological discontinuity when merging the repaired area with the existing model, this step achieves seamless 3D model updates through dynamic replacement and reconstruction of placeholder frames. This invention provides a specific embodiment: Step 105, fusing the compressed spatial volume element data, the incremental modeling processing results, and the continuous structure data to generate a real-time updated 3D model of the building complex, specifically including the following steps:

[0187] Step 501: Perform spatial occupancy mapping processing on the compressed spatial unit data to generate the basic building geometric occupancy frame.

[0188] In this step, spatial occupancy mapping processing refers to the operation of converting the geometric center coordinates of the compressed unit into spatial positioning points and establishing the topological relationship between units; the basic building geometric occupancy frame refers to the building outline skeleton formed by the occupancy points, which preserves the topological connection relationship of key structures.

[0189] In this embodiment of the invention, the geometric center coordinates of the compressed spatial volume element data are first converted into spatial occupant points. Next, topological adjacency relationships between these occupant points are established. Finally, a basic architectural geometric occupant framework representing the building foundation outline is generated.

[0190] Step 502: Input the structural change features in the incremental modeling processing results into the basic building geometric occupancy frame to generate an initial dynamically updated occupancy frame.

[0191] In this step, structural change features refer to the differential dataset output by incremental modeling, including the coordinate set of newly added units and the identifiers of units to be deleted; the initial dynamic update placeholder frame refers to the placeholder frame after fusing structural changes, reflecting the addition and deletion changes in the current scan cycle.

[0192] In this embodiment of the invention, the coordinates of newly added units and the identifiers of deleted units in the incremental modeling processing results are first parsed. Next, new unit occupant points are inserted into the basic building geometry placeholder frame, and deleted unit points are removed. Finally, an initial dynamically updated placeholder frame containing the latest structural changes is generated.

[0193] Step 503: Locate the spatial region corresponding to the continuous structural data in the initial dynamic update placeholder framework, replace the spatial region with the spatial triangular mesh structure in the continuous structural data, and generate the target dynamic update placeholder framework.

[0194] In this step, the spatial region refers to the physical spatial range in the target dynamically updated placeholder frame that matches the boundary unit of the continuous structure data; the target dynamically updated placeholder frame refers to the placeholder frame after the repair region replacement is completed, which contains the vertex information of the repair structure.

[0195] In this embodiment of the invention, firstly, the boundary cell identifiers of the continuous structural data are matched with the cell identifiers in the initial dynamically updated placeholder frame. Secondly, the spatial region to be repaired within the frame is located. Subsequently, all placeholder points within this region are deleted, and triangular mesh vertices of the continuous structural data are inserted. Finally, the repaired target dynamically updated placeholder frame is generated.

[0196] Step 504: Perform topology connection reconstruction on the target dynamically updated placeholder frame to generate a complete building topology network.

[0197] In this step, topology connection reconstruction refers to the operation of rebuilding the connection edges between the repair area and the retained area to restore the overall structural continuity; complete building topology network refers to the reconstructed spatial node connection network to ensure topological connectivity of all structural units.

[0198] In this embodiment of the invention, the adjacency relationship between newly added and retained units in the target dynamically updated placeholder framework is first detected. Secondly, the topological connection edges of the boundary points of the repair area are reconstructed. Finally, a complete building topology network with complete spatial connectivity is generated.

[0199] Step 505: Perform spatial triangulation on the spatial node coordinates of the complete building topology network to generate a set of triangular facets.

[0200] In this step, spatial node coordinates refer to the three-dimensional position data of all vertices in the topology network; spatial triangulation refers to the operation of generating triangular patches from surface nodes, while preserving the internal topological relationships; the triangular patch set refers to the triangular mesh dataset covering the building surface, containing vertex coordinates and patch connection relationships.

[0201] In this embodiment of the invention, the coordinates of all spatial nodes in the complete building topology network are first extracted. Next, triangulation is performed on the surface nodes, while the internal nodes maintain topological connectivity. Finally, a set of triangulated patches covering the building surface is generated.

[0202] Step 506: Bind the triangular facet set to the surface material features of the pre-set building complex to generate a real-time updated 3D model of the building complex.

[0203] In this step, surface material features refer to the distribution data of building facade materials, including texture images and physical attribute labels; binding operation refers to the process of mapping material data to the spatial coordinates of triangular facets to achieve the association between visual and physical attributes.

[0204] In this embodiment of the invention, firstly, the material distribution map of a pre-set historical model of a building complex is retrieved. Secondly, the spatial coordinates of the triangular facet set are aligned with the coordinates of the material map. Finally, a material pixel mapping and binding operation is performed to generate a real-time updated 3D model of the building complex.

[0205] This invention achieves seamless fusion of multi-source data through a dynamic placeholder framework, eliminates and repairs cracks in the connection between the structure and the original model by utilizing topological reconstruction, rapidly generates a visual surface based on spatial triangulation, and preserves architectural aesthetic features by combining historical material binding, and finally outputs a real-time 3D model that balances geometric accuracy and visual realism in an edge computing environment, which significantly improves the efficiency of on-site decision-making in large-scale building renovation projects.

[0206] Figure 2 This invention provides a schematic diagram of a real-time modeling system for field survey data based on edge computing, as shown in the embodiment of the invention. Figure 2 As shown, the system includes:

[0207] The acquisition module 21 is used to acquire three-dimensional spatial point data of the pre-set building complex;

[0208] The partitioning module 22 is used to perform dynamic spatial volume unit partitioning processing on the three-dimensional spatial point data based on the surface curvature characteristics of the three-dimensional spatial point data, so as to generate compressed spatial volume unit data.

[0209] Modeling module 23 is used to perform incremental modeling processing on the compressed spatial volume unit data and identify structural discontinuities from the incremental modeling processing results;

[0210] Repair module 24 is used to perform spatial connection path repair processing on the geometric topological relationship of the discontinuous region of the structure to generate continuous structural data;

[0211] The fusion module 25 is used to fuse the compressed spatial volume unit data, the incremental modeling processing results, and the continuous structure data to generate a real-time updated 3D model of the building complex.

[0212] Figure 2 The aforementioned real-time modeling system for field survey data based on edge computing can execute... Figure 1 The implementation principle and technical effects of the real-time modeling method for field survey data based on edge computing described in the illustrated embodiment will not be repeated here. The specific methods by which each module and unit of the real-time modeling system for field survey data based on edge computing performs its operations have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0213] In one possible design, Figure 2The real-time modeling system for field survey data based on edge computing, as shown in the embodiment, can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0214] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0215] The processing component 32 is used to: acquire three-dimensional spatial point data of a pre-set building complex; perform dynamic spatial unit division processing on the three-dimensional spatial point data based on the surface curvature characteristics of the three-dimensional spatial point data to generate compressed spatial unit data; perform incremental modeling processing on the compressed spatial unit data to identify structural discontinuities from the incremental modeling processing results; perform spatial connection path repair processing on the geometric topological relationships of the structural discontinuities to generate continuous structural data; and fuse the compressed spatial unit data, the incremental modeling processing results, and the continuous structural data to generate a real-time updated three-dimensional model of the building complex.

[0216] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0217] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0218] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0219] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0220] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0221] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0222] This invention also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown is a real-time modeling method for field survey data based on edge computing.

[0223] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0224] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0225] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0226] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A real-time modeling method for field survey data based on edge computing, characterized in that, include: Obtain 3D spatial point data of the pre-designed building complex; Based on the surface curvature characteristics of the three-dimensional spatial point data, the three-dimensional spatial point data is dynamically divided into spatial volume units to generate compressed spatial volume unit data. Incremental modeling is performed on the compressed spatial volume element data, and structural discontinuities are identified from the incremental modeling results. Spatial connection path repair processing is performed on the geometric topological relationships of the discontinuous regions of the structure to generate continuous structural data; The compressed spatial volume unit data, the incremental modeling processing results, and the continuous structure data are fused to generate a real-time updated 3D model of the building complex. Specifically, incremental modeling processing is performed on the compressed spatial volume element data, and structural discontinuities are identified from the incremental modeling results, including: From the compressed spatial unit data, a set of newly added spatial units and a set of historically modeled spatial units are divided. The spatial positional relationship between the newly added spatial unit set and the historically modeled spatial unit set is established to obtain a cross-period spatial adjacency table. The surface curvature compatibility parameter of each adjacent unit pair in the cross-period spatial adjacency table is calculated. The surface curvature compatibility parameter is a quantitative index of the difference in surface curvature values ​​between adjacent unit pairs. Each surface curvature compatibility parameter is quantified to generate a set of periodic compatibility measurement values. Applying the distribution characteristics of the periodic compatibility measurement value set, a dynamic compatibility threshold is set. Adjacent unit pairs whose periodic compatibility measurement values ​​are lower than the dynamic compatibility threshold are identified as abnormal adjacent unit pairs. These abnormal adjacent unit pairs are aggregated to generate an initial set of fracture unit pairs. Spatial continuity analysis is performed on the initial set of fracture unit pairs to generate structural fracture clusters. Each structural fracture cluster is mapped to the surface space of the pre-set building complex to generate structural discontinuities.

2. The method according to claim 1, characterized in that, Obtain 3D spatial point data of the pre-designed building complex, including: Collect motion trajectory data of a pre-set mobile scanning device in a pre-set building complex area, analyze the motion trajectory data, and generate a discrete trajectory point set; The three-dimensional coordinate points of each building in the pre-set building complex are collected at the spatial locations corresponding to each discrete trajectory point in the discrete trajectory point set. Establish the spatiotemporal correlation information between the discrete trajectory point set and the three-dimensional coordinate point set; Based on the spatiotemporal correlation information, each three-dimensional coordinate point in the set of three-dimensional coordinate points is time-stamped to generate a sequence of three-dimensional coordinate points with timestamps. The sequence of timestamped 3D coordinate points is divided into continuous scan segments, and multiple subsequences of 3D coordinate points with continuous scan paths are extracted from the processing results. The subsequences of the three-dimensional coordinate points are merged to generate three-dimensional spatial point data.

3. The method according to claim 1, characterized in that, Based on the surface curvature features of the three-dimensional spatial point data, dynamic spatial volume element partitioning is performed on the three-dimensional spatial point data to generate compressed spatial volume element data, including: Calculate the surface curvature value within the neighborhood of each spatial point data item in the three-dimensional spatial point data; The surface curvature values ​​are aggregated to generate a set of spatial point curvature values. Based on the numerical distribution characteristics of the set of spatial point curvature values, the pre-set building complex is divided into a high curvature region and a low curvature region. Dense spatial meshing and sparse spatial meshing are then performed on the high curvature region and the low curvature region, respectively, to generate an initial set of spatial volume unit boundaries. Based on the gradient change trend of the set of spatial point curvature values, the partition density transition parameter between adjacent spatial units in the initial set of spatial unit boundaries is adjusted to generate an optimized set of spatial unit boundaries. The partition density transition parameter refers to the adjustment coefficient that controls the rate of change of adjacent unit size. Based on the optimized set of spatial volumetric unit boundaries, the three-dimensional spatial point data is subjected to spatial unit attribution mapping to generate a spatial point volumetric unit index relationship table. Using the spatial point volume element index relationship table, calculate the geometric center coordinates of all spatial points within each spatial volume element, and generate a spatial volume element coordinate set as the compressed spatial volume element data.

4. The method according to claim 1, characterized in that, Spatial connection path repair processing is performed on the geometric topological relationships of the discontinuous regions of the structure to generate continuous structural data, including: Identify the outer contour spatial volume elements of the discontinuous regions of the structure to generate a set of fracture boundary elements; Using the spatial coordinates of each unit in the fracture boundary unit set, a topological connection skeleton is constructed as the initial connection path network; Based on the preset structural mechanics constraints, a set of candidate repair paths is generated from the initial connection path network. Calculate the spatial orientation compatibility parameters between the candidate repair path set and the set of building component units adjacent to the fracture boundary unit set, aggregate the spatial orientation compatibility parameters to generate a path compatibility metric set, wherein the spatial orientation compatibility parameter refers to the quantified value of the consistency between the repair path and the direction of the adjacent component; Based on the set of path compatibility metrics, optimized repair paths that meet preset continuity conditions are selected from the set of candidate repair paths. The optimized repair path is converted into a spatial triangular mesh structure, and the spatial triangular mesh structure is input into the discontinuous region of the structure to generate continuous structure data.

5. The method according to claim 4, characterized in that, Calculate the spatial orientation compatibility parameters between the candidate repair path set and the set of building component units adjacent to the fracture boundary unit set, aggregate the spatial orientation compatibility parameters, and generate a path compatibility metric set, including: The spatial direction vectors of each candidate repair path in the candidate repair path set and the main direction vectors of each building component in the building component set adjacent to the fracture boundary unit set are projected and calculated to generate a path component projection relationship set. The cosine of the angle between each of the spatial direction vectors and each of the principal direction vectors is calculated using the path component projection relationship set to generate a path direction compatibility value set. Detect the spatial distance deviation between the endpoints of each candidate repair path and the geometric center point of the building component unit corresponding to each endpoint, and generate a path endpoint deviation set; The set of path direction compatibility values ​​and the set of path endpoint deviations are weighted and fused to generate a single path compatibility parameter set. The single path compatibility parameters are integrated based on their path identifiers in the single path compatibility parameter set to generate a set of path compatibility metrics.

6. The method according to claim 1, characterized in that, The compressed spatial unit data, the incremental modeling results, and the continuous structure data are fused to generate a real-time updated 3D model of the building complex, including: The compressed spatial volume unit data is subjected to spatial occupancy mapping processing to generate a basic building geometric occupancy frame; The structural change features in the incremental modeling processing results are input into the basic building geometric occupancy frame to generate an initial dynamically updated occupancy frame. The structural change features refer to the difference dataset output by the incremental modeling, including the coordinate set of newly added units and the identifier of units to be deleted. In the initial dynamic update placeholder framework, locate the spatial region corresponding to the continuous structural data, replace the spatial region with the spatial triangular mesh structure in the continuous structural data, and generate the target dynamic update placeholder framework. The target dynamically updated placeholder framework is reconstructed using topology connections to generate a complete building topology network. The spatial node coordinates of the complete building topology network are spatially triangulated to generate a set of triangular facets. The triangular facet set is bound to the surface material features of the pre-set building complex to generate a real-time updated 3D model of the building complex.

7. A real-time modeling system for field survey data based on edge computing, used to execute the real-time modeling method for field survey data based on edge computing as described in any one of claims 1 to 6, characterized in that, include: The acquisition module is used to acquire three-dimensional spatial point data of a pre-set building complex; The partitioning module is used to perform dynamic spatial volume unit partitioning on the three-dimensional spatial point data based on the surface curvature features of the three-dimensional spatial point data, so as to generate compressed spatial volume unit data. The modeling module is used to perform incremental modeling processing on the compressed spatial volume unit data and identify structural discontinuities from the incremental modeling processing results. The repair module is used to perform spatial connection path repair processing on the geometric topological relationships of the discontinuous regions of the structure to generate continuous structural data; The fusion module is used to fuse the compressed spatial volume unit data, the incremental modeling processing results, and the continuous structure data to generate a real-time updated 3D model of the building complex.

8. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a real-time modeling method for field survey data based on edge computing as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The system contains a computer program that, when executed by a computer, implements a real-time modeling method for field survey data based on edge computing as described in any one of claims 1 to 6.

Images