A plate recognition method based on geometric figure range discrimination

Abstract

The present application relates to the technical field of plate identification, in particular to a plate identification method based on geometric range discrimination. Mainly including the following steps: using image acquisition equipment, laser scanner and BIM model to obtain the geometric data of the plate, including boundary contour, surface morphology, spatial position and size information; extracting the boundary and key geometric feature points of the plate through graph cut algorithm; generating a three-dimensional geometric model by using multi-scale shape analysis algorithm; using Markov random field algorithm for geometric range discrimination to identify abnormal areas; comparing the BIM model design data by using iterative closest point algorithm to generate an identification report. The present application realizes accurate identification and range discrimination of building structure plates, effectively improving the monitoring and management efficiency of construction quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of plate recognition, and in particular to a plate recognition method based on geometric figure range determination. BACKGROUND

[0002] In building structure design, plates are widely used in various buildings as key structural components for bearing and transferring loads. Plates in building structures are usually made of materials such as concrete and steel, with various shapes and sizes, and have their own geometric characteristics according to different bearing requirements and structural arrangements. However, with the complication of modern building design, especially in high-rise buildings, special-shaped buildings and complex frame structures, the geometric shapes and layouts of plates become more complex. There are still the following problems: traditional plate recognition methods mostly rely on manual visual inspection or simple mechanical recognition means, which are not only inefficient and have high error rates; traditional methods are mostly used for planar or simple two-dimensional structures, and cannot effectively recognize and process complex three-dimensional structure plates, lacking fine three-dimensional geometric model generation and analysis means, resulting in low recognition accuracy in actual application; existing technologies are easily affected by geometric shape and spatial position changes when processing complex building plates, resulting in inaccurate or unstable range determination results, which cannot meet the needs of high-precision recognition. SUMMARY

[0003] To solve the above problems, the present application provides a plate recognition method based on geometric figure range determination, which solves the problem of low efficiency and high error of plate recognition methods when the geometric shape and spatial position change in complex building design, and effectively generates and analyzes complex three-dimensional geometric models to ensure the accuracy and stability of the range determination results to meet the needs of high-precision recognition, thereby realizing accurate recognition and range determination of building structure plates.

[0004] To achieve the above purpose, the technical solution adopted by the present application is:

[0005] A plate recognition method based on geometric figure range determination, comprising the following steps:

[0006] S1: acquiring geometric data of building structure plates by using image acquisition equipment, laser scanners and BIM models; the geometric data includes boundary contours, surface morphology, spatial position and size information of building structure plates;

[0007] S2: based on the geometric data, performing boundary extraction by graph cut algorithm to obtain contour lines and boundary feature points of building structure plates, and recognizing key feature points including geometric inflection points and curve features of building structure plates;

[0008] S3: analyzing the key feature points by using a multi-scale shape analysis algorithm to generate a three-dimensional geometric model of the building structure plate;

[0009] S4: constructing a geometric figure range discrimination model based on the three-dimensional geometric model using a Markov random field algorithm to perform range discrimination, and generating a range discrimination result;

[0010] S5: based on the range discrimination result, comparing and verifying the range discrimination result identified based on the iterative closest point algorithm with design data in the BIM model, and generating an identification report of the building structure plate.

[0011] Further, the building structure plate includes a concrete plate, a steel structure plate, an aluminum plate, and a composite structure plate.

[0012] Further, the step S2 includes the following steps:

[0013] S21: based on the geometric data, collecting point cloud data of the building structure plate by a laser scanner, and performing a preprocessing operation on the point cloud data; the point cloud data includes three-dimensional coordinate point information of the building structure plate;

[0014] S22: extracting a boundary contour of the preprocessed point cloud data by a graph cut algorithm, combining boundary data of the BIM model, and determining an initial boundary of the building structure plate;

[0015] S23: based on the initial boundary, combining the material characteristics of the building structure plate, and using a normal vector smoothing algorithm to optimize the preliminary boundary and identify key feature points; the key feature points include geometric inflection points, straight edges, curved segments, and curvature change feature points of the building structure plate.

[0016] Further, the step S23 includes the following steps:

[0017] According to the geometric data and material characteristics of the building structure plate, the surface normal vector is calculated, and a preliminary normal vector field is constructed based on the distribution of the normal vector;

[0018] The normal vector field is weighted and smoothed using a weighted average method to generate an optimized normal vector field;

[0019] Using the optimized normal vector field, based on curvature change analysis and differential geometry feature extraction algorithm, the key feature points of the building structure plate are identified.

[0020] Further, the step S3 includes the following steps:

[0021] S31: based on the key feature points, constructing a preliminary geometric contour of the building structure plate;

[0022] S32: using a multi-scale shape analysis algorithm to analyze the key feature points, and generating a three-dimensional geometric model combining the preliminary geometric contour of the building structure plate;

[0023] S33: performing surface refinement processing on the generated three-dimensional geometric model, eliminating irregularities by using a smooth surface fitting algorithm, optimizing boundary precision, and obtaining an optimized three-dimensional geometric model.

[0024] Further, the step S4 includes the following steps:

[0025] S41: based on the three-dimensional geometric model, constructing a preliminary geometric range of the building structural slab, defining geometric boundary conditions and constraints of the building structural slab;

[0026] S42: constructing a geometric graph range discrimination model based on the geometric boundary conditions of the building structural slab by using a Markov random field algorithm, and modeling and optimizing the relationship between adjacent geometric units in the model; the nodes in the geometric graph range discrimination model represent geometric boundary points and feature points, and the edges represent the mutual geometric relationship between these points;

[0027] S43: optimizing the Markov random field model by using an energy function, calculating the energy value of each geometric unit, and identifying abnormal areas or areas that do not meet the design specifications in the geometric range;

[0028] S44: generating a geometric range discrimination result based on the optimized geometric graph range discrimination model, and outputting the range discrimination information of the building structural slab.

[0029] Still further, the formula of the geometric graph range discrimination model is as follows:

[0030]

[0031] wherein E(X) represents the total energy of the geometric graph range discrimination model, which is used to discriminate whether the geometric range of the building structural slab meets the specifications; V represents the node set of the geometric graph, i.e., the key feature points on the building structural slab; E represents the edge set of the geometric graph, i.e., the mutual relationship between the key feature points; N i represents the normal direction of the building structural slab surface at the feature point i; represents the normal direction of the building structural slab at the feature point i in an ideal case; x i and x j represent the position vectors of the nodes i and j, i.e., the three-dimensional coordinates of the adjacent two key feature points on the building structural slab; α represents the weight parameter of the normal vector difference; β represents the weight parameter of the geometric relationship between the adjacent feature points.

[0032] Further, the step S5 includes the following steps:

[0033] S51: Based on the geometric range discrimination result, the boundary feature information of the building structure plate is extracted, and the identified geometric feature points and boundary lines are preliminarily aligned with the design data in the BIM model;

[0034] S52: Using an iterative closest point algorithm to accurately align the recognition results, and optimizing the registration accuracy between the geometric model of the building structure plate and the BIM model design data through multiple iterations;

[0035] S53: Based on the alignment result, identifying the deviation area between the geometric model and the design data, analyzing the nature and source of the deviation, and determining whether the building structure plate has a geometric deviation exceeding the allowable tolerance;

[0036] S54: Generate an identification report of the building structure plate, the report content including geometric deviation information, spatial location of the deviation area, deviation range, and degree of conformity with the BIM model design data.

[0037] Furthermore, the formula of the iterative closest point algorithm is as follows:

[0038]

[0039] Among them, T k+1 represents the transformation matrix after the k+1th iteration; T represents the rigid body transformation matrix, which is the rotation and translation operation of the building structure plate geometric data collected by the laser scanner; p i represents the three-dimensional coordinates of the i-th feature point on the building structure plate collected by the laser scanner; q i Indicates the BIM model with p i The three-dimensional coordinates of the nearest design point corresponding to the feature point; N represents the total number of feature point pairs; || T(p i )-q i || 2 represents the squared Euclidean distance.

[0040] The beneficial effects of the present invention are:

[0041] The present application can accurately obtain the geometric data of the building structure plate by combining the image acquisition device, laser scanner and BIM model, realize the accurate description of the contour, surface morphology, spatial position and size of the plate, and provide high-precision basic data for subsequent identification. Advanced computer vision and mathematical models such as graph cut algorithm, Markov random field algorithm and multi-scale shape analysis algorithm can automatically identify the boundary, contour and feature points of the building structure plate, reduce manual intervention, and improve the identification efficiency and intelligent level. Through multi-scale shape analysis of key feature points, a three-dimensional geometric model is generated, which makes the identification process more stereoscopic and detailed, and can accurately present the morphology and position of the building structure plate, providing strong support for subsequent comparison. The geometric figure range discrimination model constructed by the Markov random field algorithm can accurately discriminate the range of the building structure plate, ensuring that the identification result is more accurate in the spatial range. The iterative closest point algorithm is used to compare and verify the identified geometric model with the design data in the BIM model, ensuring the accuracy and reliability of the identification result, effectively reducing the error, and finally generating a reliable identification report to provide data support for engineering construction and monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0042] Fig. 1 is a flowchart of a plate identification method based on geometric figure range discrimination according to the present application.

[0043] Fig. 2 is a flowchart of step S4 provided by an embodiment of the present application.

[0044] Fig. 3 is a flowchart of step S5 provided by an embodiment of the present application. DETAILED DESCRIPTION

[0045] Referring to Figs. 1-3 The present application relates to a plate identification method based on geometric figure range discrimination.

[0046] EMBODIMENT

[0047] A plate identification method based on geometric figure range discrimination includes the following steps:

[0048] S1: Obtain the geometric data of the building structure plate by using the image acquisition device, laser scanner and BIM model; the geometric data includes the boundary contour, surface morphology, spatial position and size information of the building structure plate; the building structure plate includes concrete plate, steel structure plate, aluminum plate and composite structure plate.

[0049] Specifically, image acquisition devices suitable for the construction site environment are selected, such as high-resolution digital cameras, unmanned aerial photography equipment, or stereo vision cameras. Depending on the specific environment and conditions of the building structure plate, industrial-grade cameras with dustproof and waterproof functions can be selected. For example, using a drone for high-altitude wide-range shooting is particularly suitable for large-scale building structures such as steel structure plates and concrete plates. For indoor structure plates (such as composite structure plates or steel structure plates), stereo cameras mounted on rails can be used for close-range shooting.

[0050] Laser scanners suitable for the current size and location of the building structure plate are selected. For example, using a three-dimensional laser scanner (such as a LiDAR device) for all-around scanning can provide millimeter-level accuracy, suitable for geometric data acquisition of larger or more complex structure plates. For scenarios requiring high-precision geometric information, high-precision static laser scanners can be used. This device can be fixed at a specific location and perform multiple scans on the plate's boundaries, surfaces, and spatial position information to obtain complete three-dimensional point cloud data. If the building structure plate is large in area or the environment is complex, dynamic laser scanners (such as portable laser scanning devices or LiDAR systems mounted on drones) can also be used. This device can be moved to collect data and cover the entire building structure plate through wide-range scanning.

[0051] BIM (Building Information Modeling) data is directly extracted from the design stage of the construction project. The BIM model contains design information of the building plate, such as size, boundary shape, thickness, material, and installation location. By accessing BIM software (such as Revit, AutoCAD, or other building design tools), geometric data related to the building structure plate is exported. These geometric data include two-dimensional plane information (such as the boundary lines of the plate) and three-dimensional geometric shapes (such as the surface model and spatial position of the plate).

[0052] The geometric data obtained from image acquisition devices, laser scanners, and BIM models is integrated. Image data and laser scanning point cloud data are registered in the same coordinate system using a spatial coordinate system. Data fusion technology is used to fuse data from the three sources into a complete three-dimensional geometric model. Image data provides texture and color information, laser scanning provides accurate spatial position information, and BIM model provides design reference. To ensure data accuracy, calibration plates or control points with known ground positions are used to calibrate image acquisition devices and laser scanners, reducing data deviations caused by device differences or environmental factors.

[0053] S2: Based on the geometric data, boundary extraction is performed through a graph cut algorithm to obtain the contour lines and boundary feature points of the building structure plate, and key feature points are identified, including geometric inflection points and curve features of the building structure plate;

[0054] The step S2 includes the following steps:

[0055] S21: Collect point cloud data of the building structure plate by a laser scanner based on the geometric data, and perform a preprocessing operation on the point cloud data; the point cloud data includes three-dimensional coordinate point information of the building structure plate;

[0056] S22: Extract the boundary contour of the preprocessed point cloud data by a graph cut algorithm, and determine the initial boundary of the building structure plate in combination with the boundary data of the BIM model;

[0057] Specifically, the preprocessed three-dimensional point cloud data is projected onto a two-dimensional plane to form a two-dimensional contour of the building structure plate. This is because the boundary of most building plates can be effectively extracted by its planar contour. The projected data is discretized to create an adjacency graph containing all points, and each node in the graph represents a three-dimensional point, and the edges represent the spatial relationship between the nodes.

[0058] The graph cut algorithm is applied to the two-dimensional projection of the point cloud to extract the boundary contour. The algorithm constructs a weighted graph model from the point cloud data, treats the boundary of the building plate as a cut line, and determines the optimal boundary contour by optimizing the cut line weight. The weight value in the graph cut algorithm is related to the characteristics of the point cloud data, especially based on the distance between points, the difference in surface normal vector, the boundary information of the BIM model, etc. The minimum cut method is used to find the cut line that minimizes the cost in the graph model, and the cut line is the initial boundary of the building structure plate. The initial boundary extracted by the graph cut algorithm is compared with the design boundary in the BIM model to ensure that the extracted boundary is basically consistent with the design model. If the initial boundary deviates from the boundary of the BIM model, the spatial position of the point cloud data is adjusted by matching to further correct the position and shape of the initial boundary.

[0059] S23: Based on the initial boundary, in combination with the material characteristics of the building structure plate, the normal vector smoothing algorithm is used to optimize the preliminary boundary and identify key feature points; the key feature points include geometric inflection points, straight edges, curved segments, and curvature change feature points of the building structure plate.

[0060] The step S23 includes the following steps:

[0061] According to the geometric data and material characteristics of the building structure plate, the surface normal vector is calculated, and a preliminary normal vector field is constructed based on the distribution of the normal vector;

[0062] Specifically, the surface normal vector of each point is calculated according to the three-dimensional geometric data (point cloud) and material characteristics of the building structure plate. The normal vector is a vector representing the surface direction, which is usually generated by fitting a plane in the neighborhood or by a triangular mesh. A preliminary normal vector field is constructed based on the normal vectors of all surface points of the building plate. The normal vector field can be regarded as the distribution of normal vectors of points on the surface of the building plate, which can describe the geometric shape of the plate surface. For the abrupt part (such as the edge or inflection point) in the normal vector field, special marking is needed to accurately capture the geometric changes in the subsequent feature point recognition process.

[0063] The normal vector field is weighted and smoothed by using a weighted average method to generate an optimized normal vector field;

[0064] Specifically, when smoothing the normal vector field, a weighted average method is used to smooth the local fluctuations of the normal vector of each point by weighting the normal vector, and to remove noise. In the weighted average process, the contribution of each point in the neighborhood to the normal vector of the target point is related to its distance and geometric position. The closer the point, the greater the weight, and the change of the normal vector of the point is inversely proportional.

[0065] The weighted average formula of the normal vector is as follows:

[0066]

[0067] where n'(p) represents the optimized normal vector of point p. This is the final normal vector of point p after weighted smoothing, which represents the direction of the surface of the point; p represents a sampling point in the point cloud of the building plate being processed, which is a point in the three-dimensional point cloud with three-dimensional coordinates (x p , y p , z p ) at a certain position on the surface of the building structure plate; N(p) represents the neighborhood of point p; w(p, q) represents the weight between point p and neighborhood point q. The weight value depends on the distance between point p and q, and the similarity of the normal vectors of the two. For the point cloud data of the building structure plate, if p and q are close in distance or similar in normal vector direction, the weight is larger; n(q) represents the normal vector of point q.

[0068] Using the optimized normal vector field, the key feature points of the building structure plate are identified based on curvature change analysis and differential geometry feature extraction algorithm.

[0069] Specifically, in the optimized normal vector field, the curvature is estimated by calculating the neighborhood normal vector change rate of each point. Based on the principle of differential geometry, the change of the normal vector in the neighborhood is calculated using the curvature formula to identify the area where significant bending changes occur on the surface. Common curvature calculation methods include neighborhood surface fitting or finite difference method.

[0070] The identification of feature points is as follows:

[0071] Geometric inflection point identification: By analyzing the curvature changes in the normal vector field, the geometric inflection points of the building structure are identified. Geometric inflection points are usually locations where the normal vector changes suddenly, and correspond to corners and turning points on the boundary.

[0072] Straight edge recognition: In areas where the normal vector changes gently, the straight edges of the building panels are identified by linearly fitting the normal vector. These edges are typically areas with consistent normal vector directions and zero or minimal curvature.

[0073] Curve segment identification: A curve segment is an area where the surface normal vector changes gradually along a curve. The curvature is relatively smooth but significantly different from the straight edge. By analyzing the rate of change of the normal vector's curvature, the boundaries of the curve segment can be identified.

[0074] Identification of curvature mutation points: In the normal vector field, curvature mutation points refer to areas with a large rate of curvature change, such as extreme points and turning points of a curve. These mutation points usually correspond to important geometric features of building panels.

[0075] S3: Analyze the key feature points using a multi-scale shape analysis algorithm to generate a three-dimensional geometric model of the building structure plate;

[0076] Wherein, the step S3 includes the following steps:

[0077] S31: constructing a preliminary geometric outline of the building structure plate based on the key feature points;

[0078] Specifically, a preliminary geometric outline is constructed from key feature points. These feature points are then connected through a connection algorithm to form a preliminary outline of the building board. This process includes both straight-line and curve fitting to ensure that the connecting lines between the feature points accurately reflect the overall outline of the building board.

[0079] Line Fitting: For straight edges, a least squares fitting algorithm is used to generate an optimal straight line connecting all feature points on that edge. This process eliminates noise and ensures a smooth edge.

[0080] Curve Fitting: For curved segments, polynomial curve fitting or spline interpolation algorithms are used to connect the characteristic points of the curve section to generate a smooth curve profile. Interpolation methods such as B-splines or Catmull-Rom splines are used to ensure that the generated curve is accurate near inflection points and points of curvature change.

[0081] The fitted contour lines are projected into the three-dimensional space to form a preliminary geometric framework of the building structure plate. The framework shows the main structural contours of the plate, including boundary lines, straight-line sections, curved sections, and corner points. These geometric information provides a basis for subsequent model generation. For complex-shaped building plates, such as curved plates or composite plates with concave-convex changes, this contour framework can preliminarily represent the general shape of the plate, but does not yet contain surface details.

[0082] S32: Analyzing the key feature points using a multi-scale shape analysis algorithm, and generating a three-dimensional geometric model in combination with the preliminary geometric contour of the building structure plate;

[0083] Specifically, a preliminary coarse-scale three-dimensional geometric model is generated based on the preliminary geometric contour. In this stage, the overall shape of the building plate is mainly fitted through key feature points and contour lines. This coarse model describes the general position, size, and boundary shape of the building plate. The details are gradually increased, especially the local optimization of curved sections. When dealing with curved sections, a curved surface fitting algorithm (such as Bézier surface, NURBS surface, etc.) is used to optimize the local geometric shape, ensuring smooth transition of the curved surface and accuracy of the boundary. In this stage, parts with large curvature changes are analyzed in detail to ensure the accuracy of the curve sections and curved surfaces. By processing the geometric shape in layers, the Gaussian pyramid algorithm is used to optimize the model layer by layer. The geometric information of each layer is further enhanced in the next level of processing. At each scale level, the curvature changes and normal vector field near the key points are accurately calculated and refined in combination with the material properties of the building plate (such as the roughness of the concrete plate surface, the smoothness of the aluminum plate), ensuring that the generated three-dimensional model can reflect the material characteristics of the actual building structure plate. Based on the above multi-scale shape analysis results, in combination with the preliminary geometric contour of the building plate, the final three-dimensional geometric model is generated. This model should include all key features of the plate, such as boundaries, surface curves, inflection points, and curvature change points, and highly coincide with the geometric shape of the actual building structure.

[0084] S33: Performing surface refinement processing on the generated three-dimensional geometric model, using a smooth curved surface fitting algorithm to eliminate irregularities, optimize boundary accuracy, and obtain an optimized three-dimensional geometric model.

[0085] S4: Based on the three-dimensional geometric model, a Markov Random Field algorithm is used to construct a geometric graph range discrimination model to perform range discrimination, and a range discrimination result is generated;

[0086] The step S4 includes the following steps:

[0087] S41: Based on the three-dimensional geometric model, a preliminary geometric range of the building structure plate is constructed, and the geometric boundary conditions and constraints of the building structure plate are defined;

[0088] S42: Construct a geometric range discrimination model based on the geometric boundary conditions of the building structure plate through the Markov Random Field algorithm, and model and optimize the relationship between adjacent geometric units in the model; the nodes in the geometric range discrimination model represent geometric boundary points and feature points, and the edges represent the mutual geometric relationship between these points;

[0089] Specifically, the three-dimensional geometric model of the building structure plate is represented as a Markov Random Field (MRF) model. MRF describes the relationship between geometric units through the nodes and edges of the graph. Nodes represent geometric boundary points and feature points, and edges represent the mutual geometric relationship between these points. In this model, each geometric feature point of the building structure plate (such as inflection points, boundary points, curve segments, etc.) is represented as a node. These nodes are connected by edges, and the weight of the edge represents the spatial relationship (such as distance, angle, curvature change between adjacent points, etc.) between the geometric feature points.

[0090] The relationship between adjacent geometric units is modeled through the edges in the Markov Random Field. These relationships include:

[0091] Geometric distance: the Euclidean distance between two adjacent geometric points. This is used to represent the spatial proximity between geometric feature points. The closer the distance, the higher the weight of the edge, indicating a stronger connection between the two points.

[0092] Normal vector relationship: the difference in normal vectors of two geometric feature points. By calculating the change in normal vectors of adjacent geometric units, the curvature change of the building structure plate can be described. If the normal vector difference is large, it may indicate an inflection point or a region with large curvature.

[0093] Curvature change: by comparing the curvature of adjacent geometric points, it is identified whether the surface of the building plate has undergone significant bending. This is particularly important when dealing with curved plates and composite plates with concave and convex structures.

[0094] To improve the accuracy of the geometric range discrimination model, an optimization algorithm is used to adjust the edge weights in the model. By minimizing the distance difference, normal vector difference, and curvature change between adjacent points, it ensures that the relationship between adjacent geometric units in the model accurately reflects the geometric shape of the real building plate.

[0095] S43: Use the energy function to optimize the Markov Random Field model, calculate the energy value of each geometric unit, and identify abnormal areas or areas that do not meet design specifications within the geometric range;

[0096] It should be noted that the result of energy optimization can effectively identify the area in the building board that does not meet the geometric specification. The nodes of these areas will show high energy values during the optimization process, indicating that the geometric characteristics of this area are significantly different from other parts, which may be a design or construction error. Abnormal areas may include areas where the edge exceeds the design range, parts of the surface with uneven concave and convex surfaces, or areas with uneven thickness.

[0097] S44: Based on the optimized geometric range determination model, a geometric range determination result is generated, and range determination information of the building structure board is output.

[0098] Further, the formula of the geometric range determination model is as follows:

[0099]

[0100] wherein E(X) represents the total energy of the geometric range determination model, which is used to determine whether the geometric range of the building structure board meets the specification; V represents a node set of the geometric figure, i.e., a key feature point on the building structure board; E represents an edge set of the geometric figure, i.e., the mutual relationship between the key feature points; N i represents the normal direction of the surface of the building structure board at the feature point i; represents the normal direction of the building structure board at the feature point i in an ideal case; x i and x j represent the position vectors of nodes i and j, i.e., the three-dimensional coordinates of two adjacent key feature points on the building structure board; a represents a weight parameter of the normal vector difference, which is used to control the contribution of the normal vector deviation to the total energy; and b represents a weight parameter of the geometric relationship between adjacent feature points, which controls the influence of the distance deviation between the feature points on the total energy.

[0101] S5: Based on the range determination result, the range determination result identified based on the iterative closest point algorithm is compared and verified with the design data in the BIM model, and an identification report of the building structure board is generated;

[0102] The step S5 includes the following steps:

[0103] S51: Based on the geometric range determination result, the boundary feature information of the building structure board is extracted, and the identified geometric feature points and boundary lines are preliminarily aligned with the design data in the BIM model;

[0104] S52: The iterative closest point algorithm is used to accurately align the identification result, and the registration accuracy of the geometric model of the building structure board and the design data in the BIM model is optimized through multiple iterations;

[0105] Specifically, the Iterative Closest Point (ICP) algorithm is employed to further refine the alignment between the geometric model and the BIM model. The ICP algorithm works by finding the closest point pairs between two sets of point cloud data (the geometric model and the design data) and minimizing the distance between them in each iteration. The goal of the ICP algorithm is to spatially align the geometric range discrimination results with the design data in the BIM model through multiple iterations.

[0106] The specific steps of the ICP algorithm include: for each feature point in the geometric model, find the closest point in the BIM design data to form a point pair. Based on the distance between the point pairs, calculate the rigid transformation parameters of the geometric model, including translation and rotation. Through least squares method, the best rigid transformation is calculated to minimize the distance between the two sets of points. Apply the calculated transformation to the geometric model to update the position and pose of the model. Repeat the above steps to gradually reduce the distance between the two sets of points until the set error threshold or the iteration limit is reached. After each iteration, the alignment effect is evaluated by calculating the root mean square error of the two sets of data. When the error is less than the predetermined threshold, it indicates that the alignment result meets the accuracy requirements.

[0107] S53: According to the alignment result, identify the deviation area between the geometric model and the design data, analyze the nature and source of the deviation, and determine whether the building structure plate has geometric deviation beyond the allowed tolerance;

[0108] It should be noted that based on the ICP alignment result, the deviation between each point in the geometric model and the corresponding point in the BIM design data is calculated. The deviation includes the difference in geometric parameters such as plane position, thickness, surface curvature, etc. Use the deviation analysis tool to identify the areas in the geometric model that deviate greatly from the design data. Common deviation areas include boundary line misalignment, uneven thickness, abnormal surface undulation, etc.

[0109] The analysis of the nature of the deviation is as follows:

[0110] Geometric deviation classification: classify the geometric deviation into different types, such as:

[0111] Boundary deviation: the boundary of the building plate does not match the design boundary, such as edge misalignment or bending.

[0112] Thickness deviation: the actual thickness does not match the design thickness, which may cause structural performance problems.

[0113] Surface morphology deviation: deviation in surface flatness, which may affect surface smoothness or strength.

[0114] Deviation size calculation: for each deviation area, calculate the specific numerical value of the deviation, including distance deviation (millimeters), curvature deviation, etc., to ensure that the degree of each deviation area can be quantified.

[0115] Based on the tolerance requirements in the design specifications and BIM model, each deviation area is determined to see if it exceeds the permissible tolerance range. Areas outside the tolerance range are marked as unacceptable and require processing or correction. If the deviation is within the permissible range, the geometric model is considered to meet the design requirements.

[0116] S54: Generate an identification report of the building structure plate, the report content including geometric deviation information, spatial location of the deviation area, deviation range, and degree of conformity with the BIM model design data.

[0117] Furthermore, the formula of the iterative closest point algorithm is as follows:

[0118]

[0119] Among them, T k+1 represents the transformation matrix after the k+1th iteration; T represents the rigid body transformation matrix, which is the rotation and translation operation of the building structure plate geometric data collected by the laser scanner; p i represents the three-dimensional coordinates of the i-th feature point on the building structure plate collected by the laser scanner; q i Indicates the BIM model with p i The three-dimensional coordinates of the nearest design point corresponding to the feature point; N represents the total number of feature point pairs; || T(p i )-q i || 2 Represents the square of the Euclidean distance, which is used to measure the distance between the characteristic point of the i-th building structure plate after transformation T and the nearest design point q in the BIM model. i The geometric deviation between them.

[0120] In summary, this invention, through the fusion of geometric data from image acquisition devices, laser scanners, and BIM models, leverages the strengths of these different devices to ensure comprehensive and accurate geometric information. Using a graph cut algorithm and a normal vector smoothing algorithm, it efficiently extracts the boundaries of building structural panels and identifies key geometric features, such as inflection points, straight edges, and curved segments. This process, combined with the design data from the BIM model, effectively optimizes boundary locations and ensures high-precision restoration of geometric features.

[0121] This invention utilizes a multi-scale shape analysis algorithm to optimize the geometric model layer by layer, accurately reproducing complex building panel forms. This approach is particularly suitable for panels with large surface curvature or uneven structures. Furthermore, refinement of curvature and normal vectors ensures smooth model transitions and boundary accuracy. Range discrimination based on a Markov random field model automatically identifies anomalous areas within the geometric range, particularly geometric deviations that may arise during construction. Energy optimization and modeling of the spatial relationships between geometric elements ensure the accuracy of the discrimination results.

[0122] The present application adopts an iterative closest point algorithm (ICP) to accurately align a geometric model and BIM design data, and can identify deviations between the geometric model and design specifications. The finally generated identification report records the geometric deviations and their positions in detail, and provides an important basis for subsequent quality control and correction.

[0123] The above embodiments only describe the preferred embodiments of the present application, and do not limit the scope of the present application. Without departing from the design spirit of the present application, various modifications and improvements to the technical solutions of the present application made by ordinary engineering technicians in the art shall fall within the protection scope determined by the claims of the present application.

Claims

1. A plate recognition method based on geometric figure range discrimination, characterized in that: The following steps are involved: S1: using an image acquisition device, a laser scanner, and a BIM model to acquire geometric data of a building structural plate; the geometric data includes boundary contours, surface morphology, spatial position, and size information of the building structural plate; S2: Based on the geometric data, performing boundary extraction using a graph cut algorithm to obtain the contour lines and boundary feature points of the building structure plate, and identifying key feature points, including geometric inflection points and curve features of the building structure plate; S3: Analyze the key feature points using a multi-scale shape analysis algorithm to generate a three-dimensional geometric model of the building structure plate; S4: Based on the three-dimensional geometric model, a Markov random field algorithm is used to construct a geometric figure range discrimination model for range discrimination and generate range discrimination results; S5: Based on the range identification result, the range identification result identified by the iterative closest point algorithm is compared and verified with the design data in the BIM model, and an identification report of the building structure plate is generated; The step S2 comprises the following steps: S21: Based on the geometric data, collecting point cloud data of the building structure plate by using a laser scanner, and performing a pre-processing operation on the point cloud data; the point cloud data includes three-dimensional coordinate point information of the building structure plate; S22: Extract boundary contours from the pre-processed point cloud data using a graph cut algorithm, and determine the initial boundary of the building structure plate by combining it with the boundary data of the BIM model; S23: Based on the initial boundary and in combination with the material properties of the building structure plate, the initial boundary is optimized using a normal vector smoothing algorithm to identify key feature points; the key feature points include geometric inflection points, straight line edges, curve segments, and curvature change feature points of the building structure plate; The step S23 includes the following steps: According to the geometric data and material properties of the building structure plate, its surface normal vector is calculated, and a preliminary normal vector field is constructed based on the distribution of the normal vector; Performing weighted smoothing processing on the normal vector field using a weighted average method to generate an optimized normal vector field; Using the optimized normal vector field, the key feature points of the building structure plate are identified based on curvature change analysis and differential geometry feature extraction algorithm; The step S3 comprises the following steps: S31: constructing a preliminary geometric outline of the building structure plate based on the key feature points; S32: Analyze the key feature points using a multi-scale shape analysis algorithm, and generate a three-dimensional geometric model based on the preliminary geometric outline of the building structure plate; S33: performing surface refinement processing on the generated three-dimensional geometric model, eliminating irregularities using a smooth surface fitting algorithm, optimizing boundary accuracy, and obtaining an optimized three-dimensional geometric model; The step S4 comprises the following steps: S41: Based on the three-dimensional geometric model, construct a preliminary geometric range of the building structure plate, and define geometric boundary conditions and restriction conditions of the building structure plate; S42: constructing a geometric figure range discrimination model based on the geometric boundary conditions of the building structure plate using a Markov random field algorithm, and modeling and optimizing the relationships between adjacent geometric units in the model; the nodes in the geometric figure range discrimination model represent geometric boundary points and feature points, and the edges represent the mutual geometric relationships between these points; S43: Optimize the Markov random field model using the energy function, calculate the energy value of each geometric unit, and identify abnormal areas or areas that do not meet the design specifications within the geometric range; S44: Based on the optimized geometric range discrimination model, a geometric range discrimination result is generated, and range discrimination information of the building structure plate is output.

2. A plate recognition method based on geometric figure range discrimination according to claim 1, characterized in that: The building structural panels include concrete panels, steel structural panels, aluminum panels and composite structural panels.

3. The plate recognition method based on geometric figure range discrimination according to claim 1, characterized in that: The formula of the geometric figure range discrimination model is as follows: ； in, It represents the total energy of the geometric range discrimination model, which is used to determine whether the geometric range of the building structure plate meets the specifications; V represents the node set of the geometric figure, that is, the key feature points on the building structure plate; E represents the edge set of the geometric figure, that is, the relationship between the key feature points; Indicates the normal direction of the building structure plate surface at the feature point i; represents the normal direction of the building structure plate at the characteristic point i under ideal conditions; and Represents the position vectors of nodes i and j, i.e., the three-dimensional coordinates of two adjacent key feature points on the building structure plate; The weight parameter representing the difference of the normal vector; The weight parameter that represents the geometric relationship between adjacent feature points.

4. The plate recognition method based on geometric figure range discrimination according to claim 1, characterized in that: The step S5 comprises the following steps: S51: Based on the geometric range discrimination result, the boundary feature information of the building structure plate is extracted, and the identified geometric feature points and boundary lines are preliminarily aligned with the design data in the BIM model; S52: Using an iterative closest point algorithm to accurately align the recognition results, and optimizing the registration accuracy between the geometric model of the building structure plate and the BIM model design data through multiple iterations; S53: Based on the alignment result, identifying the deviation area between the geometric model and the design data, analyzing the nature and source of the deviation, and determining whether the building structure plate has a geometric deviation exceeding the allowable tolerance; S54: Generate an identification report of the building structure plate, the report content including geometric deviation information, spatial location of the deviation area, deviation range, and degree of conformity with the BIM model design data.

5. The plate recognition method based on geometric figure range discrimination according to claim 4 is characterized in that: The formula of the iterative closest point algorithm is as follows: ； in, represents the transformation matrix after the k+1th iteration; T represents the rigid body transformation matrix, that is, the rotation and translation operations on the geometric data of the building structure plate collected by the laser scanner; represents the three-dimensional coordinates of the i-th feature point on the building structure plate collected by the laser scanner; Indicates the BIM model The three-dimensional coordinates of the corresponding nearest design point; N represents the total number of feature point pairs; represents the squared Euclidean distance.

Images