Construction method of site model based on BIM technology

Abstract

The application relates to the technical field of data processing, and relates to a site field model construction method based on BIM technology, which comprises the following steps: acquiring at least two images and at least three feature points, each image is divided into a construction area and a non-construction area; determining a target local geometric structure by using the at least three feature points, determining the credibility weight of a scene structure, and then determining the measured coordinates of a target feature point; determining the correction weight of the target feature point in a target construction area by using the measured coordinates and a construction completion coefficient of the target construction area, and correcting BIM design coordinates; determining the deviation correction value of a target non-construction area by using the measured coordinates, original BIM design coordinates and a foundation treatment quality coefficient; correcting an initial BIM model by using the corrected BIM design coordinates and the deviation correction value, and constructing a site to obtain construction data. The application effectively eliminates the deviation between a theoretical model and actual construction, improves model precision, and reduces construction errors.

Description

Technical Field

[0001] This invention relates to the field of data processing, and more specifically to a method for constructing a site model based on BIM technology. Background Technology

[0002] In the field of architectural engineering, Building Information Modeling (BIM) technology has become an important tool for improving project management efficiency. Through 3D visualization modeling, BIM technology can intuitively present the spatial relationships of site topography, building layout, and the surrounding environment, effectively overcoming the limitations of traditional 2D drawings in terms of information representation. During construction, various professional teams can collaborate based on a unified BIM model, resolving issues such as material conflicts and site layout in real time, significantly reducing rework rates. Furthermore, the various engineering data integrated into the BIM model can provide dynamic simulation support for the construction process, improving the scientific validity and feasibility of construction plans.

[0003] However, traditional BIM model building has significant limitations. These models are primarily based on theoretical parameters established during the design phase, while the actual conditions at the construction site are constantly changing. Over time, the discrepancies between the theoretical model and the actual site accumulate, potentially leading to significant spatial location differences. This discrepancy is particularly pronounced when temporary additions to buildings are required, potentially severely impacting the positioning accuracy and construction quality of these new structures. Existing technologies lack effective dynamic model correction mechanisms, making it difficult to eliminate accumulated errors in a timely manner and meet the demands of real-time changes at the construction site. Especially when dealing with the differences between construction and non-construction areas, existing methods often employ a uniform correction strategy, failing to fully consider the different characteristics of the two types of areas, resulting in discrepancies between the correction results and the actual situation.

[0004] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0005] To address the existing technical problems, the present invention aims to provide a site model construction method based on BIM technology, which can effectively improve model progress and, by utilizing the optimized model for construction design, reduce construction errors.

[0006] The specific technical solution adopted is as follows: A method for constructing a site model based on BIM technology is provided, including: acquiring at least two images and acquiring at least three feature points from the at least two images, and dividing each image into a construction area and an unconstructed area, wherein the at least two images are images of the same site acquired at different times; using the at least three feature points to determine the target local geometric structure, thereby determining the credibility weight of the scene structure corresponding to the at least two images, and then determining the measured coordinates of each target feature point; using the measured coordinates and the construction completion coefficient of the target construction area to determine the correction weight of each target feature point in the target construction area, thereby determining the corrected BIM design coordinates of the target construction area; using the measured coordinates, the original BIM design coordinates, and the foundation treatment quality coefficient of the target unconstructed area to determine the deviation correction value of the target unconstructed area; using the corrected BIM design coordinates and the deviation correction value to correct the initial BIM model, and using the corrected BIM model to construct the site and obtain construction data.

[0007] In one embodiment of the present invention, the step of acquiring at least two images, acquiring at least three feature points in the at least two images, and dividing each image into a construction area and an unconstructed area includes: acquiring images of the target site using an image acquisition device to acquire at least two images; acquiring multiple feature points in each image, and acquiring at least three feature points adjacent to each target feature point; acquiring the same stable feature points in the at least two images, and determining the construction area and the unconstructed area based on the stable feature points.

[0008] In one embodiment of the present invention, the step of determining the target local geometry using at least three feature points to determine the credibility weight of the scene structure corresponding to the at least two images, and then determining the measured coordinates of each target feature point, includes: obtaining the target local geometry corresponding to at least three feature points adjacent to the target feature point, and determining the angle error of the target feature point using the target local geometry; obtaining the average deviation of the target feature point from the initial design angle in the at least two images; determining the credibility weight of the scene structure corresponding to the at least two images using the angle error and the average deviation; and determining the measured coordinates of each target feature point using the credibility weight and the coordinate information of the at least two images.

[0009] In one embodiment of the present invention, the step of obtaining the target local geometric structure corresponding to at least three feature points adjacent to the target feature point, and determining the angle error of the target feature point based on the target local geometric structure, includes: obtaining a first angle of the target local geometric structure in a first image, and obtaining a second angle of the target local geometric structure in a second image, wherein the at least two images include at least a first image and a second image; determining the angle error of the target feature point using the first angle and the second angle; and determining the measured coordinates of each target feature point using the confidence weight and the coordinate information of the at least two images, including: obtaining the first coordinate information of the target feature point in the first image, and obtaining the second coordinate information of the target feature point in the second image; and performing a weighted calculation using the confidence weight, the first coordinate information, and the second coordinate information to determine the measured coordinates of the target feature point.

[0010] In one embodiment of the present invention, determining the correction weight of each target feature point in the target construction area using the measured coordinates and the construction completion coefficient of the target construction area, and then determining the corrected BIM design coordinates of the target construction area, includes: obtaining the BIM design coordinates, construction completion coefficient, and maximum allowable deviation of the target construction area; determining the distance deviation using the measured coordinates and the BIM design coordinates; obtaining the total number of feature points that meet preset requirements within a preset range; determining the correction weight of each target feature point in the target construction area using the total number of feature points, the distance deviation, the BIM design coordinates, the construction completion coefficient, and the maximum allowable deviation; and determining the corrected BIM design coordinates of the target construction area using the correction weight, the maximum allowable deviation, the measured coordinates, and the BIM design coordinates.

[0011] In one embodiment of the present invention, determining the corrected BIM design coordinates of the target construction area using the corrected weight, the maximum allowable deviation, the measured coordinates, and the BIM design coordinates includes: determining a deviation adjustment coefficient using the maximum allowable deviation; and determining the corrected BIM design coordinates of the target construction area using the corrected weight, the deviation adjustment coefficient, the measured coordinates, and the BIM design coordinates.

[0012] In one embodiment of the present invention, determining the deviation correction value of the target unconstructed area using the measured coordinates, the original BIM design coordinates, and the foundation treatment quality coefficient of the target unconstructed area includes: obtaining the foundation treatment quality coefficient, the original soil area stability index coefficient, and the confidence level of the target unconstructed area; and determining the deviation correction value of the target unconstructed area using the measured coordinates, the original BIM design coordinates, the foundation treatment quality coefficient, the original soil area stability index coefficient, and the confidence level.

[0013] In one embodiment of the present invention, obtaining the foundation treatment quality coefficient, the original soil area stability index coefficient, and the reliability of the target unconstructed area includes: obtaining the regional reflectance standard deviation and the regional design threshold of the target unconstructed area; determining the compaction gradient weight using the regional reflectance standard deviation and the regional design threshold; determining the foundation treatment quality coefficient using the compaction gradient weight; obtaining the surface roughness and geological stratification reliability of the target unconstructed area; determining the original soil area stability index coefficient using the surface roughness and the geological stratification reliability; obtaining the measured point cloud density and the designed point cloud data standard information of the target unconstructed area; and determining the reliability using the time decay factor, the measured point cloud density, and the point cloud data standard information.

[0014] In one embodiment of the present invention, determining the foundation treatment quality coefficient using the compaction gradient weight includes: obtaining a first measured compaction degree at the edge of the foundation zone of the target unconstructed area, obtaining a second measured compaction degree at the edge of the original soil zone of the target unconstructed area, and obtaining the total width of the transition zone of the target unconstructed area, wherein the transition zone exists between the foundation zone and the original soil zone; determining the transition zone effect weight using the first measured compaction degree, the second measured compaction degree, and the total width; and determining the foundation treatment quality coefficient using the transition zone effect weight and the compaction gradient weight.

[0015] In one embodiment of the present invention, the step of correcting the initial BIM model using the corrected BIM design coordinates and the deviation correction value, and then constructing the site using the corrected BIM model to obtain construction data, includes: constructing an initial BIM model; correcting the initial BIM model using the corrected BIM design coordinates and the deviation correction value to obtain a corrected BIM model; aligning the newly added building information with the measured coordinates corresponding to the corrected BIM model, and then inputting the aligned newly added building information into the corrected BIM model to construct the site to obtain construction data for the newly added building information.

[0016] The beneficial effects of this invention are as follows: It provides a method for constructing a site model based on BIM technology, comprising: acquiring at least two images and at least three feature points from the at least two images, and dividing each image into a construction area and an unconstructed area, wherein the at least two images are images acquired at different times for the same site; determining the target local geometry using the at least three feature points to determine the credibility weight of the scene structure corresponding to the at least two images, and then determining the measured coordinates of each target feature point; determining the correction weight of each target feature point in the target construction area using the measured coordinates and the construction completion coefficient of the target construction area, and then determining the corrected BIM design coordinates of the target construction area; determining the deviation correction value of the target unconstructed area using the measured coordinates, the original BIM design coordinates, and the foundation treatment quality coefficient of the target unconstructed area; correcting the initial BIM model using the corrected BIM design coordinates and the deviation correction value, and constructing the site using the corrected BIM model to obtain construction data. In this invention, by dynamically correcting the coordinate parameters of the construction area and the unconstructed area in the BIM model, combined with multi-time-point image feature analysis, the cumulative deviation between the theoretical model and the actual construction is effectively eliminated, effectively improving model accuracy and reducing construction errors. Attached Figure Description

[0017] To more clearly illustrate the technical solutions and advantages 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 only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the site model construction method based on BIM technology provided by the present invention. Detailed Implementation

[0019] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a site model construction method based on BIM technology proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0021] In existing technologies, traditional BIM model building methods rely on theoretical parameters from the design phase, making it difficult to cope with dynamic changes on the construction site. As construction progresses, deviations between the theoretical model and the actual site gradually accumulate. This deviation can lead to inaccurate positioning, especially when temporary additions are needed, impacting construction efficiency. For example, when a construction team plans to add temporary facilities near a completed area, the original model may experience coordinate shifts due to a lack of consideration for foundation settlement or differences in construction progress, resulting in conflicts between the new building and the existing structure.

[0022] To address the aforementioned issues, this application provides a site model construction method based on BIM technology. It proposes using multi-time-point image data and feature point analysis to distinguish between construction and non-construction areas. Furthermore, by calculating the confidence weights and correction parameters for different areas, the model coordinates are adjusted accordingly, thereby accurately reflecting the dynamically changing site conditions in the model.

[0023] The following description, in conjunction with the accompanying drawings, details a specific scheme for a site model construction method based on BIM technology provided by the present invention.

[0024] Please see Figure 1 The diagram illustrates a flowchart of the site model construction method based on BIM technology provided by the present invention.

[0025] like Figure 1 As shown, the method for constructing a site model based on BIM technology includes the following steps:

[0026] S10. Acquire at least two images and at least three feature points from the at least two images, and divide each image into a construction area and an unconstructed area, wherein the at least two images are images of the same site acquired at different times.

[0027] Among them, at least two images refer to overall images of the construction site, which may include construction areas and non-construction areas; feature points refer to points with significant characteristics, such as column corners, wall edges, etc. It can be understood that feature points appearing in both images can be stable feature points; construction areas refer to areas where construction has begun, and non-construction areas refer to areas where construction has not yet begun.

[0028] Specifically, an image acquisition device is used to acquire multiple images of the construction site at different time points, and at least three feature points that exist in all multiple images are acquired; and the images are divided into regions to distinguish between construction areas and non-construction areas.

[0029] S20. Use at least three feature points to determine the local geometric structure of the target, so as to determine the confidence weight of the scene structure corresponding to at least two images, and then determine the measured coordinates of each target feature point.

[0030] Among them, the target local geometry refers to the region enclosed by the lines connecting at least three feature points; the scene structure refers to the scene structure that appears in at least two images, such as the target local geometry; the confidence weight refers to the confidence weight value corresponding to the feature point; the measured coordinates refer to the real coordinates of the feature point in the same construction coordinate system; the target feature point refers to the currently selected feature point, which can be understood as each feature point being used as the target feature point when it is selected.

[0031] Specifically, after acquiring at least three feature points, the at least three feature points are connected to form a line, and the local geometric structure of the target is determined by the line connecting the at least three feature points. The target geometric structure in at least two images is then compared to obtain the confidence weight of the scene structure corresponding to at least two images. The confidence weight is then used to calculate the measured coordinates of each target feature point in the same construction coordinate system in at least two images.

[0032] S30. Using the measured coordinates and the construction completion coefficient of the target construction area, determine the correction weight of each target feature point in the target construction area, and then determine the corrected BIM design coordinates of the target construction area.

[0033] Among them, the target construction area refers to the currently selected construction area, and each construction area can be used as the target construction area when it is selected; the construction completion coefficient refers to the progress indicator of construction in the target construction area; the correction weight refers to the weight required for subsequent coordinate correction of the construction area; and the corrected BIM design coordinates refer to the coordinates after the original BIM design coordinates have been corrected by the correction weight.

[0034] Specifically, after obtaining the measured coordinates, the construction completion coefficient of the target construction area is obtained. Then, using the measured coordinates and the construction completion coefficient of the target construction area, the correction weight of the target feature points of the target construction area is determined. Then, using the measured coordinates, the original BIM actual coordinates and the correction weight, the corrected BIM design coordinates of the target construction area are determined.

[0035] S40. Using the measured coordinates, the original BIM design coordinates, and the foundation treatment quality coefficient of the target unconstructed area, determine the deviation correction value of the target unconstructed area.

[0036] Among them, the foundation treatment quality coefficient refers to the parameter used to quantitatively evaluate the foundation treatment effect in untreated areas. For example, it can be achieved by calculating the ratio of the regional reflectance standard deviation to the regional design threshold, combined with the transition zone effect weight. This coefficient can reflect the uniformity and compaction difference of foundation treatment.

[0037] Specifically, after obtaining the foundation treatment quality coefficient of the target unconstructed area, the deviation correction value of the feature points of the current target unconstructed area is calculated using the obtained measured coordinates, original BIM design coordinates, and foundation treatment quality coefficient.

[0038] S50. Correct the initial BIM model using the corrected BIM design coordinates and deviation correction values, and construct the site using the corrected BIM model to obtain construction data.

[0039] The initial BIM model is an existing model related to the construction site design.

[0040] Specifically, after obtaining the corrected BIM design coordinates and deviation correction values, the model is corrected by using the corrected BIM design coordinates and deviation correction values ​​to correspond to the initial BIM model, thus obtaining the corrected BIM model; then, the corrected BIM model is used to construct the overall construction site, including the construction area and the non-construction area, thereby obtaining the construction data of the entire site, so that the construction data can be used to carry out construction on the entire construction site.

[0041] In this embodiment, compared with existing technologies, traditional methods rely on static design parameters and do not consider dynamic changes and regional differences during construction. This solution effectively reduces cumulative deviations by using multi-time-point image data and a regional correction mechanism. Simultaneously, it utilizes the characteristic differences between construction and non-construction areas to specifically adjust model parameters, significantly improving the accuracy of new building positioning. Through the above technical solution, this application can correct deviations between the BIM model and the actual site conditions in real time. Especially in the case of temporarily added buildings, by adjusting regional coordinates, it avoids positioning errors caused by construction progress or foundation changes, thereby reducing rework and improving construction efficiency.

[0042] Furthermore, acquiring at least two images, acquiring at least three feature points from the at least two images, and dividing each image into a construction area and an unconstructed area may also include the following operations.

[0043] An image acquisition device is used to acquire images of the target site to obtain at least two images.

[0044] Obtain multiple feature points in each image, and obtain at least three feature points adjacent to each target feature point.

[0045] Obtain the same stable feature points in at least two images, and use the stable feature points to determine the construction area and the unconstructed area.

[0046] Image acquisition devices refer to equipment used to capture site images. Specifically, this can be achieved using drones equipped with multispectral cameras or LiDAR scanners, ensuring complete site coverage through multi-angle imaging. Stable feature points are feature points that maintain consistent positions in images taken at different times. These can be achieved using corner points of fixed structures within the site or high-reflectivity markers, utilizing their spatial stability as a benchmark for area division. Adjacent feature points are sets of points geometrically related to the target feature point. Specifically, feature matching algorithms can be used to select the three points closest to the target point, forming a triangular topology to support local geometric analysis.

[0047] Specifically, the image acquisition device periodically photographs the site along a preset flight path, for example, collecting orthophotos and oblique photography data weekly. During feature point extraction, the SIFT algorithm is used to identify key points in the images, and a geometric association group containing three adjacent points is established for each target feature point. For stable feature point selection, a bidirectional matching algorithm is used to compare the coordinates of feature points in images from different times, retaining strong feature points with matching errors less than a threshold and consistent reflection intensity. The division between construction and non-construction areas is based on the spatial distribution of stable feature points; for example, areas containing concrete structures or completed foundation treatment are classified as construction areas, while exposed soil or untreated areas are classified as non-construction areas.

[0048] For example, image acquisition devices such as drones need to collect flight data across the entire construction area. Before flight, flight paths must be planned based on the building structures of the constructed areas and the foundation areas of the unconstructed areas. To eliminate errors caused by variations in the drone's own parameters and ensure that subsequent data deviations only stem from environmental factors, the drone's camera shooting parameters (such as flight altitude, camera angle, and sensor parameters) must strictly adhere to the settings from the BIM modeling phase, while ensuring that the overlap rate between the flight path and the lateral direction is no less than 80%. During flight, multi-dimensional data such as reflectivity and elevation of the shooting area are recorded simultaneously. After data collection, noise reduction and stitching processes are performed to generate a complete 3D point cloud model.

[0049] In this embodiment, accurate division between construction and non-construction areas was achieved, providing a reliable foundation for subsequent zoning and correction of the BIM model. The zoning method based on stable feature points can adapt to dynamic site changes and eliminate the influence of temporary objects on construction status assessment. The geometrical correlation between adjacent feature points enhances the robustness of feature point selection, avoiding zoning errors caused by anomalies in a single feature point.

[0050] In some embodiments, the target local geometry is determined using at least three feature points to determine the confidence weight of the scene structure corresponding to at least two images, and then the measured coordinates of each feature point are determined. The following operations may also be included.

[0051] Obtain the target local geometry corresponding to at least three feature points adjacent to the target feature point, and determine the angle error of the target feature point based on the target local geometry.

[0052] Obtain the average deviation of the target feature points from the initial design angle in at least two images.

[0053] By utilizing angular error and average deviation, the credibility weights of the scene structure corresponding to at least two images are determined.

[0054] Using confidence weights and coordinate information from at least two images, the measured coordinates of each target feature point are determined.

[0055] The target local geometry refers to the spatial geometric relationship formed by adjacent feature points. Specifically, it can be represented by the angle or side length ratio formed by the lines connecting feature points, reflecting the stability of the local structure. Angle error refers to the angular difference of the same local geometric structure in different images. This can be achieved by calculating the difference between corresponding angles in the first and second images, used to assess geometric distortion caused by environmental interference or equipment errors during image acquisition. Average deviation refers to the degree of deviation of a target feature point from its theoretically designed angle in multiple images. This can be achieved by statistically analyzing the average difference between the measured and designed angle values ​​of the feature point in multiple images, used to quantify the positional offset of the feature point in the actual scene. Confidence weight is a parameter calculated based on angle error and average deviation to measure the reliability of image coordinate data. This can be implemented using weighted averaging or error propagation models, used to assign different confidence levels to data from different images during coordinate fusion.

[0056] Specifically, when determining the measured coordinates of a target feature point, at least three adjacent feature points are first extracted from multiple images, and a triangle or polygon structure formed by these feature points is constructed as a local geometric reference. For example, when adjacent feature points form a triangle, the angle error can be obtained by calculating the difference between the interior angles of the triangle in the first image and the corresponding interior angles in the second image. Further, by comparing the measured angles of the target feature point in each image with the preset design angles in the BIM model, the average angle deviation of the feature point in images at different time points is calculated. Subsequently, the angle error and the average deviation are normalized and input into a preset weight calculation model to generate a reliability weight value reflecting the reliability of the image data. Finally, the coordinate data of the same feature point in different images are weighted and fused. For example, the data with coordinates (X1, Y1) and a weight of 0.7 in the first image is superimposed with the data with coordinates (X2, Y2) and a weight of 0.3 in the second image to obtain the measured three-dimensional coordinates of the feature point.

[0057] In this embodiment, the reliability of image data across multiple time periods can be dynamically evaluated, and measurement errors caused by construction progress or environmental factors can be automatically corrected during coordinate fusion. This technology effectively solves the problem of accumulated deviations caused by directly using unverified coordinate data during traditional BIM model updates, providing a high-precision feature point positioning benchmark for the zoning correction of construction and non-construction areas. For example, when locating a new building, the measured coordinates obtained by this method can control the positioning accuracy error of the corrected model within the centimeter range, ensuring the accuracy of the connection between the new structure and the existing building.

[0058] In some embodiments, obtaining the target local geometry corresponding to at least three feature points adjacent to the target feature point, and determining the angle error of the target feature point based on the target local geometry, includes:

[0059] The target local geometry is obtained at a first angle in a first image and at a second angle in a second image, wherein at least two images include at least the first image and the second image;

[0060] The angular error of the target feature point is determined using the first angle and the second angle.

[0061] Using confidence weights and coordinate information from at least two images, the measured coordinates of each target feature point are determined, including:

[0062] Obtain the first coordinate information of the target feature points in the first image, and obtain the second coordinate information of the target feature points in the second image;

[0063] The measured coordinates of the target feature points are determined by weighting the confidence weight, the first coordinate information, and the second coordinate information.

[0064] The target local geometry refers to the spatial geometric relationship formed by at least three adjacent feature points, which can be realized using triangulation or polygon topology to reflect the relative positional relationship between feature points. The first angle and the second angle refer to the angular differences of the same local geometric structure in images taken at different times. This can be achieved by extracting geometric edges and calculating the included angle using image processing algorithms, quantifying the geometric deformation caused by changes in shooting perspective. The angle error refers to the angular difference of the same geometric structure in different images, which can be achieved by calculating the absolute value or standard deviation of the two angle differences, used to assess the impact of environmental interference on the stability of the geometric structure during image acquisition. The confidence weight is a scene structure reliability index calculated based on the angle error and average deviation, which can be achieved using the reciprocal of the normalized error or an exponential decay function, used to dynamically adjust the contribution of different images in coordinate calculation. The first coordinate information and the second coordinate information refer to the two-dimensional pixel coordinates or three-dimensional spatial coordinates of the same feature point in different images, which can be achieved using feature matching algorithms or stereo vision technology, used to provide positional data from multiple perspectives. Weighted calculation refers to the linear combination of different coordinate information according to the confidence weight. Specifically, it can be implemented by weighted average or least squares optimization algorithm, and is used to fuse multi-source data and eliminate measurement errors of single images.

[0065] Specifically, after acquiring first and second images of the same location taken at different times, the local geometric structure formed by at least three neighboring feature points is extracted for each target feature point. The contour features of this geometric structure in both images are extracted using an edge detection algorithm, and the corresponding first and second angles are calculated. The difference between the two angles is used as the angle error to quantify the degree of deformation of the local structure in the two images. Simultaneously, the average deviation of the target feature point from the initial design angle in the two images is calculated, and a confidence weight is generated based on the angle error. This weight reflects the measurement reliability of the image in the current scene. Subsequently, the first coordinate information of the target feature point in the first image and the second coordinate information in the second image are extracted, and the two coordinate information are weighted and fused according to the confidence weight. For example, when the confidence weight of the first image is 0.7 and that of the second image is 0.3, the formula for calculating the measured coordinates can be expressed as: Measured coordinates = 0.7 × first coordinate + 0.3 × second coordinate. This dynamic weight allocation mechanism can automatically reduce the influence of images with large angle errors on the final coordinate calculation results.

[0066] For example, the initial BIM construction and the current drone flight data are denoted as datasets D1 and D2, respectively. That is, at least two images are the first image and the second image, with the first image being D1 and the second image being D2. For feature points that exist in both flights (such as column corners and wall edges), the nearest neighbor matching algorithm and spatial consistency check are used to retain points that satisfy existence and structural consistency and are spatially stable.

[0067] The geometric features of the feature points extracted from the two images exhibit high stability. Therefore, by calculating the angular error of the local geometric structure of the feature points in the two datasets, the reliability of the feature points can be quantified. Furthermore, by combining the design parameters in the BIM model as weight constraints: only feature points that simultaneously satisfy the conditions of small angular error in both images and small angular deviation from the BIM design have the highest reliability and should therefore have a higher weight. The formula for calculating the scene structure reliability weight is as follows:

[0068]

[0069] In the formula, Represents the current feature point Angle error for each feature point Take three adjacent feature points around it to form a local geometric structure, and calculate the angular error of this structure in the two sets of data: ,in The angle of this local structure in D1, The angle of this local structure in D2; Representing feature points Average deviation between the angle of the two photographs and the BIM design angle: ,in From the perspective of BIM design; n represents the total number of feature points.

[0070] By using a weighted average of the coordinates of two sets of data, we obtain... The actual coordinates:

[0071]

[0072] in, For feature points The true X-axis coordinates, For feature points The true Y-axis coordinate, For feature points The true Z-axis coordinates, Feature points in the first image X-axis coordinates Feature points in the first image Y-axis coordinate, Feature points in the first image Z-axis coordinate, Feature points in the second image X-axis coordinates Feature points in the second image Y-axis coordinate, Feature points in the second image The Z-axis coordinate.

[0073] By identifying common stable feature points in the two datasets The data from both datasets is converted to the same construction coordinate system. By utilizing stable feature points present in both drone images, environmental factors such as temporary facilities and environmental interference (e.g., drone positioning drift) can be accurately filtered out, ensuring that only core structural information that is stable over the long term at the construction site is processed.

[0074] This embodiment effectively solves the problem of coordinate calculation deviation caused by dynamic scene changes during multi-time period image data fusion. By establishing a reliability assessment mechanism based on the stability of local geometric structures, the calculation accuracy of measured coordinates of feature points is significantly improved while retaining the advantages of multi-source data, providing more reliable spatial reference data for subsequent BIM model correction.

[0075] In some embodiments, the correction weight of each target feature point in the target construction area is determined using measured coordinates and the construction completion coefficient of the target construction area, thereby determining the corrected BIM design coordinates of the target construction area. This may include the following operations:

[0076] Obtain the BIM design coordinates, construction completion coefficient, and maximum allowable deviation of the target construction area;

[0077] Determine the distance deviation using measured coordinates and BIM design coordinates;

[0078] Obtain the total number of target feature points that meet preset requirements within a preset range;

[0079] The correction weight of each target feature point in the target construction area is determined by using the total number of feature points, distance deviation, BIM design coordinates, construction completion coefficient, and maximum allowable deviation.

[0080] By using the corrected weights, the maximum allowable deviation, the measured coordinates, and the BIM design coordinates, the corrected BIM design coordinates of the target construction area are determined.

[0081] The construction completion coefficient is a quantitative indicator reflecting the current completion status of the construction area. It can be implemented using progress data collected by sensors or an image recognition-based area coverage assessment algorithm, and is used to measure the impact of the actual completion level of the construction area on coordinate correction. The maximum allowable deviation refers to the coordinate offset threshold allowed in the design specifications, which can be implemented through a numerical range set by engineering standards or project requirements, and is used to constrain the deviation adjustment range during the correction process. The total number of feature points refers to the number of feature points that meet the stability requirements within a preset spatial range. It can be implemented using image matching algorithms or point cloud clustering analysis to count the number of valid feature points in neighboring areas, and is used to assess the measurement reliability of local areas.

[0082] Specifically, within the construction area, the difference between BIM design coordinates and measured coordinates is calculated using distance deviation. This deviation, combined with the construction completion coefficient, reflects the impact of construction progress on coordinate stability. For example, when the construction completion coefficient is high, the constraint effect of completed structures on feature points is enhanced, and the correction weight tends to reduce the adjustment range of measured coordinates. The total number of feature points is used to assess the measurement consistency of local areas. When the number of valid feature points in adjacent areas is large, it indicates that the measurement data of that area has high reliability, and the correction weight will be dynamically adjusted according to this number. The maximum allowable deviation serves as a constraint condition to ensure that the corrected coordinates do not exceed the allowable error range of the project. By combining the correction weight with the deviation adjustment coefficient, the corrected BIM design coordinates are finally calculated, ensuring that the model adjustment in the construction area conforms to both measured data and engineering specifications.

[0083] In some embodiments, spatial coherence also needs to be obtained, and then the correction weight of each target feature point in the target construction area is determined by using spatial coherence, total number of feature points, distance deviation, BIM design coordinates, construction completion coefficient and maximum allowable deviation.

[0084] In this embodiment, the present application can dynamically adjust the coordinate correction weight according to the real-time status of the construction area, effectively suppressing the spread of deviations caused by uneven construction progress or local measurement errors while ensuring model accuracy, and providing a BIM model data foundation that matches the actual height of the site for the positioning of new buildings.

[0085] In some embodiments, the corrected BIM design coordinates of the target construction area are determined using corrected weights, maximum allowable deviation, measured coordinates, and BIM design coordinates, including:

[0086] The deviation adjustment coefficient is determined using the maximum allowable deviation.

[0087] By using correction weights, deviation adjustment coefficients, measured coordinates, and BIM design coordinates, the corrected BIM design coordinates of the target construction area are determined.

[0088] The maximum allowable deviation refers to the upper limit of the permitted positional deviation in the construction area, which can be determined by construction specifications or design requirements. It limits the range of coordinate adjustments during the correction process, preventing excessive correction from causing the model to deviate from actual construction conditions. The deviation adjustment coefficient is an adjustment parameter generated based on the maximum allowable deviation. It can be calculated using the reciprocal or exponential function of the maximum allowable deviation and is used to dynamically adjust the correction weight ratio between measured and design coordinates, ensuring that the corrected coordinates do not exceed the allowable construction range. The correction weight refers to the degree of influence of the target feature point on the final coordinate correction. It can be calculated using the total number of feature points, distance deviation, and construction completion coefficient. It reflects the stability of the construction status around the feature point, thus prioritizing areas with high reliability during the correction process. Measured coordinates refer to the actual measured coordinates on site obtained through image feature point matching and weighted calculation. This can be achieved using the weighted average of multiple image feature point coordinates and reflects the true spatial location of the construction area. BIM design coordinates refer to the theoretical coordinates preset in the original model, which can be obtained from design drawings or model parameters. They are used to compare with measured coordinates to evaluate construction deviations.

[0089] Specifically, when determining the corrected BIM design coordinates, a deviation adjustment coefficient is calculated based on the maximum allowable deviation. For example, when the maximum allowable deviation is 5 cm, the deviation adjustment coefficient can be 0.2, the reciprocal of the maximum allowable deviation. Then, a correction weight is used to dynamically adjust the measured coordinates and BIM design coordinates. For example, a weighted average formula is used to proportionally merge the measured coordinates and design coordinates. During this process, the correction weight is determined based on the total number of feature points around a feature point and the distance deviation. For example, when there are more feature points around a feature point that meet the preset requirements, the correction weight increases, and the proportion of the measured coordinates increases. The final generated corrected BIM design coordinates retain the original design intent while incorporating on-site measured data, and the deviation adjustment coefficient ensures that the correction range remains within the allowable range.

[0090] For example, the environment may differ between two shooting sessions due to the addition or removal of temporary facilities or local changes in terrain. However, buildings in the constructed area have stable edges and angles, and these stable structural features (such as concrete column corners) are key bases for correction. By extracting and matching the positions of these points (excluding temporary points that only appear once) from the two UAV point cloud data, deviation correction is performed based on these fixed geometric structures. In the unconstructed area, there are foundations and untreated original soil areas. The correction of these areas relies on surface or shallow points that reflect the regional attributes, mainly feature points marked with reflectivity in the UAV point cloud. The level of reflectivity indirectly reflects the degree of soil compaction, and the more compacted, the higher the weight. In the transition area between the two, the attributes of the points gradually change with the distance from the edge of the foundation. The reflectivity of points closer to the foundation area is close to the foundation standard, while the reflectivity of points closer to the original soil area gradually reflects the characteristics of the original soil, and its reflectivity needs to be multiplied by a linear attenuation coefficient. The points in the untreated original soil area are surface elevation points, which are determined by the flatness of the surface (calculated by the standard deviation of the elevation of the point cloud neighborhood).

[0091] The geometry of completed structures within the construction zone is fixed and will not deform, serving as the core basis for revising the BIM model. When these completed structures are constructed according to specifications, deviations from the BIM design should be controlled within acceptable limits. Furthermore, completed, fixed building sections do not exist in isolation; they form a continuous whole with surrounding completed structures that also comply with construction specifications. Isolated points are likely temporary elements, such as scaffolding that has not yet been dismantled. Final revisions will primarily refer to those completed, compliant structures that are integrated with the surrounding structures.

[0092] The area is divided according to the construction progress of the existing project stages. The areas are different in size and not necessarily connected in location for the two types of areas (constructed and unconstructed). Therefore, the constructed areas are numbered according to the construction time sequence, and there are multiple feature point areas in each constructed area.

[0093] Based on this, the first The first of the already constructed areas Correction weight of each point for:

[0094]

[0095] In the formula, This indicates the construction completion coefficient (1 for completed projects, and 0.3~0.5 for incomplete projects). It represents the distance deviation between the actual measurement and the design, calculated using the spatial distance formula between the on-site measured coordinates and the BIM design coordinates; The maximum allowable deviation is indicated by the specification (as specified in industry standards or design documents). Indicates spatial coherence, used to quantify the degree of connection with surrounding structures: Where N is the total number of neighboring structural points around the target point, and K is the number of these points that simultaneously satisfy A≥0.8 and The number of points, This refers to the average value corresponding to the existence of multiple maximum deviations.

[0096] After calculating the corrected weight W based on the structural characteristics of the constructed area, the coordinate correction formula for the constructed area is as follows:

[0097]

[0098] In the formula, This indicates the corrected BIM coordinates. Represents the BIM coordinates of the original design; Indicates the measured coordinates; This represents the deviation adjustment term, which avoids excessive correction of out-of-specification points. It takes the value of 1 for compliant deviations and decreases as the deviation increases for out-of-specification points.

[0099] In this embodiment, the coordinate correction range of the construction area can be effectively controlled to avoid construction rework caused by exceeding the allowable deviation. At the same time, by dynamically integrating measured data and design data, the consistency between the corrected BIM model and the actual construction status on site is improved, providing a reliable basis for the accurate positioning of new buildings.

[0100] In some embodiments, the deviation correction value of the target unconstructed area is determined using measured coordinates, original BIM design coordinates, and foundation treatment quality coefficient of the target unconstructed area, which may include the following operations;

[0101] Obtain the foundation treatment quality coefficient, original soil area stability index coefficient, and reliability of the target unconstructed area;

[0102] By using measured coordinates, original BIM design coordinates, foundation treatment quality coefficient, original soil area stability index coefficient, and reliability, the deviation correction value for the target unconstructed area is determined.

[0103] The foundation treatment quality coefficient is a parameter used to quantitatively evaluate the effectiveness of foundation treatment in untreated areas. Specifically, it is calculated by dividing the regional reflectance standard deviation by the regional design threshold, combined with the transition zone effect weight. This coefficient reflects the uniformity of foundation treatment and differences in compaction. The original soil stability index coefficient is a parameter used to measure the stability of the undisturbed soil in untreated areas. Specifically, it is calculated by weighting surface roughness and the reliability of geological stratification. This coefficient reflects the reliability of surface morphology changes and geological structure. The reliability level is a parameter used to characterize the reliability of measured data in untreated areas. Specifically, it is calculated by comparing the measured point cloud density with the design point cloud data standard information, combined with a time decay factor adjustment. This parameter reflects the decay effect of data acquisition quality over time.

[0104] Specifically, when determining the deviation correction value for unconstructed areas, the uniformity of foundation treatment is first assessed using the foundation treatment quality coefficient. For example, a large standard deviation in regional reflectivity indicates a gradient difference in compaction, requiring adjustment of the foundation treatment quality coefficient based on the transition zone effect weight. Secondly, the matching degree between surface roughness and geological stratification is analyzed using the original soil area stability index coefficient. For instance, the stability index coefficient needs to be reduced in areas with significant surface undulations. Subsequently, time decay compensation is applied to the measured point cloud data based on credibility. For example, when the measured point cloud density is lower than the design standard, a time decay factor is introduced to reduce credibility. Finally, the above parameters are comprehensively calculated with the measured coordinates and the original BIM design coordinates. For example, a weighted average or error propagation model is used to obtain the deviation correction value for unconstructed areas, thereby eliminating model deviations caused by uneven foundation treatment or changes in geological conditions.

[0105] In this embodiment, the model deviation problem caused by uneven foundation treatment, insufficient stability of the original soil area, or failure of data acquisition in the unconstructed area can be effectively solved. For example, under soft soil foundation or complex geological conditions, by dynamically adjusting the deviation correction value, the corrected BIM model can be ensured to be highly matched with the actual geological conditions on site, thereby providing a reliable basis for the positioning of new buildings and reducing the risk of rework caused by model deviation during construction.

[0106] In some embodiments, obtaining the foundation treatment quality coefficient, original soil stability index coefficient, and reliability of the target unconstructed area includes:

[0107] Obtain the regional reflectance standard deviation and regional design threshold of the target unconstructed area, and use the regional reflectance standard deviation and regional design threshold to determine the compaction gradient weight;

[0108] The foundation treatment quality coefficient is determined by using the compaction gradient weight;

[0109] To obtain the surface roughness and geological stratification reliability of the target undeveloped area;

[0110] The stability index coefficient of the original soil area was determined by using the surface roughness and the reliability of geological stratification.

[0111] Obtain the measured point cloud density of the target unconstructed area, as well as the designed point cloud data standard information;

[0112] The reliability level is determined by using the time decay factor, measured point cloud density, and standard information of point cloud data.

[0113] Among them, the regional reflectance standard deviation refers to the standard deviation calculated after collecting surface reflectance data through lidar or remote sensing equipment. Specifically, multispectral imaging technology can be used to obtain the reflectance distribution of different bands, and then the standard deviation can be calculated using statistical methods to reflect the spatial variability of surface compaction. The compaction gradient weight refers to a weighting coefficient generated based on the ratio of the reflectance standard deviation to a design threshold. Specifically, the standard deviation can be mapped to a preset interval through normalization, and the weight can be dynamically adjusted in conjunction with the threshold setting to quantify the impact of compaction variation trends in different regions on the quality of foundation treatment. Surface roughness refers to the surface elevation variation parameter obtained through 3D scanning or stereophotogrammetry. Specifically, surface curvature or elevation variance can be calculated using point cloud data to characterize the complexity of the surface morphology in unconstructed areas. Geological stratification confidence refers to the confidence index of the stratification structure generated based on geological exploration data. Specifically, the matching degree assessment between borehole sampling and ground-penetrating radar data can be used to reflect the accuracy of geological stratification. Measured point cloud density refers to the number of points per unit area in point cloud data acquired through laser scanning. Specifically, it can be achieved by dividing the data into spatial grids and counting the points within each grid, used to assess the completeness and accuracy of data acquisition. The time decay factor is a decay coefficient generated based on the interval between the data acquisition time and the current time. Specifically, it can be used to simulate the decreasing trend of data reliability over time using an exponential function, used to dynamically adjust the weight of historical data.

[0114] Specifically, when determining the foundation treatment quality coefficient, the reflectance distribution of the unconstructed area is first obtained using multispectral imaging technology. The standard deviation of the regional reflectance is calculated and compared with a preset regional design threshold to generate a compaction gradient weight. For example, when the standard deviation exceeds the threshold, it indicates a significant difference in compaction in the area, and the gradient weight is lowered to reflect the potential risk of foundation treatment quality. Next, combining surface roughness and geological stratification reliability, a weighted algorithm is used to calculate the stability index coefficient of the original soil area. Areas with higher surface roughness will have their stability score lowered, while areas with higher geological stratification reliability will have their score higher. When determining the reliability, the integrity of data collection is evaluated by comparing the measured point cloud density with the standard information of the designed point cloud data, and the reliability value is dynamically adjusted in conjunction with a time decay factor. For example, if the measured point cloud density is only 70% of the standard value and the data collection time is more than 30 days ago, the reliability will be lowered.

[0115] In this embodiment, the quality of foundation treatment, the stability of the original soil area, and the reliability of data in unconstructed areas can be quantified more accurately, thus providing a reliable basis for BIM model correction. For example, in areas with a high standard deviation of reflectivity, reducing the compaction gradient weight can avoid misjudgment of foundation treatment quality; in areas with abrupt changes in surface roughness, dynamically adjusting the stability index coefficient of the original soil area can provide early warning of potential collapse risks; and by combining the time decay factor with the adjustment of reliability, model deviations caused by the use of outdated data can be effectively avoided, ultimately ensuring that the corrected BIM model highly matches the actual site conditions.

[0116] In some embodiments, determining the foundation treatment quality coefficient using compaction gradient weights may include the following operations:

[0117] Obtain the first measured compaction degree of the edge of the foundation zone of the unconstructed area of ​​the target, obtain the second measured compaction degree of the edge of the original soil zone of the unconstructed area of ​​the target, and obtain the total width of the transition zone of the unconstructed area of ​​the target, wherein there is a transition zone between the foundation zone and the original soil zone;

[0118] The transition zone effect weights are determined using the first measured compaction degree, the second measured compaction degree, and the total width.

[0119] The foundation treatment quality coefficient is determined by using the transition zone effect weight and the compaction gradient weight.

[0120] The compaction gradient weight is a parameter reflecting the trend of compaction variation between the foundation zone and the original soil zone. It can be calculated as the ratio of the regional reflectance standard deviation to the regional design threshold, used to quantify the degree of difference in compaction between different zones. The transition zone effect weight is a parameter reflecting the influence of the transition zone on the foundation treatment quality. It can be calculated by dividing the difference between the first and second measured compaction degrees by the total width, used to characterize the attenuation characteristics of compaction differences in spatial distribution. The foundation treatment quality coefficient is an index comprehensively reflecting the foundation treatment effect. It can be calculated as a weighted sum of the transition zone effect weight and the compaction gradient weight, used to assess the stability of the foundation in untreated areas.

[0121] Specifically, the first measured compaction degree at the edge of the foundation zone can be obtained through on-site sampling or ground-penetrating radar detection, for example, by setting a detection point 0.5 meters away from the foundation boundary. The second measured compaction degree at the edge of the original soil zone can be measured symmetrically at 0.5 meters away from the original soil zone boundary, forming comparative data with the foundation zone. The total width of the transition zone can be determined through geological profile maps or 3D point cloud data analysis, for example, by extracting the area where the compaction degree is between the critical values ​​of the foundation zone and the original soil zone. In the calculation of the transition zone effect weight, the compaction degree difference reflects the degree of change in soil density caused by the foundation treatment, while the total width is used to standardize the spatial influence range. Finally, the foundation treatment quality coefficient, by superimposing the transition zone effect weight and the compaction degree gradient weight, can simultaneously consider the influence of local compaction degree differences and regional transition characteristics on foundation quality.

[0122] For example, the unconstructed areas on site include the foundation area, the untreated original soil area, and the transition zone between the two. The characteristics of these areas collectively determine the reliability of the correction. The compaction degree of the foundation area reflects its stability; the higher the compaction degree, the better the area has met the conditions for directly bearing the construction standards of the superstructure, and the higher the correction weight. The stability of the untreated original soil area is reflected in the flatness of the surface; the flatter the original soil area, the less disturbed it is by human activity, and the more stable it is, allowing for direct excavation according to the design. Considering these factors, the more stable areas within the unconstructed area are given a larger proportion in deviation repair, corresponding to the... The first unconstructed area Deviation correction value at each point The calculation formula is:

[0123]

[0124] In the formula, Indicates the foundation treatment quality coefficient: ,in For the compaction gradient weights: , The standard deviation of regional reflectance (obtained from UAV point cloud data). is the threshold for regional design (obtained from BIM), is the weight of the transition zone effect, represents the stability index of the original soil area, is the credibility of the unconstructed area.

[0125] The transition zone is a gradually changing boundary area between the foundation area and the original soil area. On the side close to the foundation area, the transition zone is mainly composed of foundation treatment materials, with only a small amount of original soil mixed in, and the degree of compaction is close to the standard of the foundation area. On the side close to the original soil area, it is mainly composed of original soil, with mixed materials, and the degree of compaction is much lower than the standard of the foundation. Therefore, when the transition zone is close to the core of the foundation area, the adjustment effect is weak; the closer it is to the original soil area, the stronger the adjustment effect (weakening its interference with the correction result). Then the weight of the transition zone effect The formula is:

[0126]

[0127] In the formula, represents the measured compaction degree at the edge of the foundation area, that is, the compaction degree at the starting point d = 0 of the transition zone; represents the compaction degree at the edge of the original soil area, located at the end point d = L of the transition zone, and L is the total width of the transition zone, which is determined by the construction design.

[0128] When d = 0, it is located at the starting point of the transition zone, close to the foundation area, the weight is the highest; when d = L, it is located at the end point of the transition zone, close to the original soil area, , reflecting the characteristic of low compaction in the original soil area; when located at the middle position of 0 < d < L, the weight decays with d, reflecting the gradual change characteristic of the transition zone from the foundation to the original soil.

[0129] Represents the stability index of the original soil area: , where is the surface roughness of the original soil area (obtained by calculating the standard deviation of the neighborhood elevation from point cloud data), is the credibility of geological stratification, obtained from the ratio of the measured reflectivity gradient to the designed reflectivity gradient, reflecting whether the original soil in the area conforms to the geological stratification in the design; F represents the credibility of the data in the unconstructed area.

[0130] Since the construction state of the unconstructed area is not yet fixed and may continue to change with the construction process (for example, the original soil area in the morning may become a replacement filling area in the afternoon, and the surface compaction degree continuously increases with mechanical operations). This characteristic of constantly new state leads to the inconsistency between the collected measured data (such as UAV point cloud) and the latest actual situation. Therefore, it is also necessary to consider the influence of the time interval between the collected data and the used data on the correction. Then the credibility of the data in the unconstructed area The formula is:

[0131]

[0132] In the formula, This represents the measured point cloud density. Standards for representing point cloud data in a design Time decay factor: τ is the status update cycle of the unconstructed area (which can be obtained from the construction log; for example, if there are 10 status changes in the past month, then τ = 30 / 10 = 3 days), and t is the time when using measured data. The data collection time represents the actual data acquisition time. The longer (larger) the interval between data acquisition and use, the lower its weight decreases with exponential decay.

[0133] In the formula for calculating the deviation correction value The overall stability reliability of the foundation area and the original soil area can be simplified to: ; The denominator represents the reliable measured deviation. To balance the reliability of the normalized stable region with the measured reliability, it can be simplified to: To balance the impact of both factors on the correction results. When the foundation area has high compaction and the original soil area is level, Increasing the size of the region enhances the reliability of the stable region within the molecule; when the measured data is new and accurate... The proportion of G in the molecule increases because reliable measurements better reflect the conditions at the construction site; the influence of the transition zone is reflected in G: when close to the foundation area, G is less affected by the transition zone (still maintaining a high value); when close to the original soil area, G decreases due to the mixing of materials in the transition zone. As the molecular weight decreases, the proportion of stable regions in the molecule naturally decreases.

[0134] In this embodiment, the foundation treatment quality of unconstructed areas can be accurately quantified, providing dynamic parameter support for BIM model correction, effectively reducing model coordinate deviation caused by uneven foundation settlement, and ensuring consistency between newly added building positioning data and actual on-site conditions.

[0135] In some embodiments, the initial BIM model is corrected using the corrected BIM design coordinates and deviation correction values, and the site is constructed using the corrected BIM model to obtain construction data. This may include the following operations:

[0136] Build the initial BIM model;

[0137] The initial BIM model is corrected by using the corrected BIM design coordinates and deviation correction values ​​to obtain the corrected BIM model;

[0138] The newly added building information is aligned with the measured coordinates corresponding to the corrected BIM model. Then, the aligned new building information is input into the corrected BIM model to construct the site and obtain the construction data of the new building information.

[0139] The BIM design coordinate correction refers to the optimized coordinates obtained by calculating the deviation between the measured coordinates of the construction area and the original design coordinates. Specifically, this can be achieved by dynamically correcting the coordinates using a weighted average algorithm combined with the construction completion coefficient, thus eliminating model deviations caused by differences in construction progress within the construction area. The deviation correction value refers to the adjustment parameters calculated based on the foundation treatment quality coefficient and the original soil stability index for the unconstructed area. This can be achieved through dynamic adjustment using a multi-source data fusion algorithm combined with measured point cloud density, compensating for model errors caused by changes in geological conditions in the unconstructed area. The initial BIM model refers to the three-dimensional site model constructed based on theoretical parameters from the design phase. This can be achieved using parametric modeling with BIM modeling software, serving as the baseline model for correction operations. New building information refers to the design parameters and spatial positioning data of temporarily added buildings or structures. This can be obtained through 3D scanning or manual input, and is used to generate the spatial layout of the new buildings in the corrected model.

[0140] Specifically, the initial BIM model is first constructed as a baseline model including site topography and existing structures. The BIM design coordinates are corrected by dynamically adjusting the geometric coordinates of completed construction sections in the model through the fusion of measured data from the construction area and design parameters; for example, the measured coordinates of the concrete structure are weighted and fused with the design coordinates. Deviation correction values ​​are based on foundation testing data and geological stability indicators from unconstructed areas, compensating for and correcting the foundation elevation in these areas of the model; for example, the predicted foundation settlement value is adjusted according to the compaction gradient weight. The corrected BIM model then uses a coordinate alignment module to match the new building information with the measured coordinate system; for example, a point cloud registration algorithm is used to spatially align the design coordinates of the new building with the site reference points, ultimately generating construction guidance data containing the precise location of the new building.

[0141] In this embodiment, the above-mentioned technical solution effectively solves the construction conflict problem caused by model deviation when locating new buildings. A regional correction strategy improves the matching accuracy between the model and the actual site conditions. A measured coordinate alignment mechanism ensures seamless integration of the spatial positioning of new buildings with existing structures, avoiding the risk of rework due to coordinate deviations. The correction method based on dynamic weight adjustment can adapt to changes in construction progress and fluctuations in geological conditions, providing a reliable three-dimensional spatial reference for temporary new buildings.

[0142] For example, based on the revised BIM model, the design coordinates, foundation extent, and load requirements of the new building are extracted. Through model coordinate system calibration, the design coordinate system of the new building is aligned with the measured coordinate system of the revised BIM.

[0143] The key focus is on verifying the compatibility between the new building foundation and the stable area of ​​the modified foundation: using the compaction gradient data of the foundation area, the flatness classification of the original soil area, and the boundary effect coefficient of the transition zone stored in the model, the model automatically screens whether the new building foundation is located in a high-stability area. If the foundation edge involves the transition zone (d>0.5), the model will trigger an early warning (suggesting that the foundation treatment range needs to be expanded to cover the transition zone); if the foundation is completely located in the core of the foundation area (compaction degree ≥95% of the design value) or the flat area of ​​the original soil area (roughness σ≤0.05), the foundation placement is deemed compliant.

[0144] The BIM components of the new building are implanted into the correction model according to the verified coordinates, and the foundation parameters are synchronously linked. The parameters are automatically written into the model attribute library to provide a basis for the foundation state for subsequent structural mechanics analysis.

[0145] In summary, this invention effectively eliminates the cumulative deviation between the theoretical model and the actual construction by dynamically correcting the coordinate parameters of the construction area and the non-construction area in the BIM model, combined with multi-time point image feature analysis, thereby effectively improving the model accuracy and reducing construction errors.

[0146] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0147] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for constructing a site model based on BIM technology, characterized in that, The method includes: Acquire at least two images and at least three feature points from the at least two images, and divide each image into a construction area and an unconstructed area, wherein the at least two images are images of the same site acquired at different times; The local geometry of the target is determined by using at least three feature points to determine the confidence weight of the scene structure corresponding to the at least two images, and then the measured coordinates of each target feature point are determined. Using the measured coordinates and the construction completion coefficient of the target construction area, the correction weight of each target feature point in the target construction area is determined, and then the corrected BIM design coordinates of the target construction area are determined. Using the measured coordinates, the original BIM design coordinates, and the foundation treatment quality coefficient of the target unconstructed area, the deviation correction value of the target unconstructed area is determined; The initial BIM model is corrected using the corrected BIM design coordinates and the deviation correction value, and the site is constructed using the corrected BIM model to obtain construction data. The determination of the measured coordinates of each target feature point includes: acquiring the target local geometry corresponding to at least three feature points adjacent to the target feature point, and determining the angle error of the target feature point based on the target local geometry; acquiring the average deviation of the target feature point from the initial design angle in the at least two images; determining the credibility weight of the scene structure corresponding to the at least two images using the angle error and the average deviation; and determining the measured coordinates of each target feature point using the credibility weight and the coordinate information of the at least two images.

2. The site model construction method based on BIM technology according to claim 1, characterized in that, The process of acquiring at least two images, acquiring at least three feature points from the at least two images, and dividing each image into a construction area and an unconstructed area includes: An image acquisition device is used to acquire images of the target site to obtain at least two images; Obtain multiple feature points in each of the images, and obtain at least three feature points adjacent to each target feature point; Obtain the same stable feature points in at least two images, and use the stable feature points to determine the construction area and the unconstructed area.

3. The site model construction method based on BIM technology according to claim 1, characterized in that, The step of acquiring the target local geometry corresponding to at least three feature points adjacent to the target feature point, and determining the angle error of the target feature point based on the target local geometry, includes: The target local geometry is obtained at a first angle in a first image and at a second angle in a second image, wherein the at least two images include at least the first image and the second image; The angular error of the target feature point is determined using the first angle and the second angle. The step of determining the measured coordinates of each target feature point using the confidence weight and the coordinate information of the at least two images includes: Obtain the first coordinate information of the target feature point in the first image, and obtain the second coordinate information of the target feature point in the second image; The measured coordinates of the target feature point are determined by weighting the confidence weight, the first coordinate information, and the second coordinate information.

4. The site model construction method based on BIM technology according to claim 1, characterized in that, The step of determining the correction weight of each target feature point in the target construction area using the measured coordinates and the construction completion coefficient of the target construction area, and then determining the corrected BIM design coordinates of the target construction area, includes: Obtain the BIM design coordinates, construction completion coefficient, and maximum allowable deviation of the target construction area; The distance deviation is determined using the measured coordinates and the BIM design coordinates; Obtain the total number of target feature points that meet preset requirements within a preset range; The correction weight of each target feature point in the target construction area is determined by using the total number of feature points, the distance deviation, the BIM design coordinates, the construction completion coefficient, and the maximum allowable deviation. Using the corrected weights, the maximum allowable deviation, the measured coordinates, and the BIM design coordinates, the corrected BIM design coordinates of the target construction area are determined.

5. The site model construction method based on BIM technology according to claim 4, characterized in that, The step of determining the corrected BIM design coordinates of the target construction area using the corrected weights, the maximum allowable deviation, the measured coordinates, and the BIM design coordinates includes: The deviation adjustment coefficient is determined using the maximum allowable deviation. The corrected BIM design coordinates of the target construction area are determined using the corrected weights, the deviation adjustment coefficients, the measured coordinates, and the BIM design coordinates.

6. The method for constructing a site model based on BIM technology according to claim 1, characterized in that, The step of determining the deviation correction value of the target unconstructed area using the measured coordinates, the original BIM design coordinates, and the foundation treatment quality coefficient of the target unconstructed area includes: Obtain the foundation treatment quality coefficient, original soil area stability index coefficient, and reliability of the target unconstructed area; Using the measured coordinates, the original BIM design coordinates, the foundation treatment quality coefficient, the original soil area stability index coefficient, and the confidence level, the deviation correction value of the target unconstructed area is determined.

7. The site model construction method based on BIM technology according to claim 6, characterized in that, The acquisition of the foundation treatment quality coefficient, original soil stability index coefficient, and reliability of the target unconstructed area includes: Obtain the regional reflectance standard deviation and regional design threshold of the target unconstructed area, and use the regional reflectance standard deviation and regional design threshold to determine the compaction gradient weight; The foundation treatment quality coefficient is determined using the compaction gradient weight. Obtain the surface roughness and geological stratification reliability of the target unconstructed area; The stability index coefficient of the original soil area is determined by using the surface roughness and the reliability of the geological stratification. Obtain the measured point cloud density of the target unconstructed area, as well as the designed point cloud data standard information; The reliability is determined by using the time decay factor, the measured point cloud density, and the point cloud data standard information.

8. The method for constructing a site model based on BIM technology according to claim 7, characterized in that, The step of determining the foundation treatment quality coefficient using the compaction gradient weight includes: The first measured compaction degree of the foundation zone edge of the target unconstructed area is obtained, the second measured compaction degree of the original soil zone edge of the target unconstructed area is obtained, and the total width of the transition zone of the target unconstructed area is obtained, wherein the transition zone exists between the foundation zone and the original soil zone; The transition zone effect weight is determined using the first measured compaction degree, the second measured compaction degree, and the total width. The foundation treatment quality coefficient is determined using the transition zone effect weight and the compaction gradient weight.

9. The method for constructing a site model based on BIM technology according to claim 1, characterized in that, The process involves correcting the initial BIM model using the corrected BIM design coordinates and the deviation correction value, and then constructing the site using the corrected BIM model to obtain construction data, including: Build the initial BIM model; The initial BIM model is corrected using the corrected BIM design coordinates and the deviation correction value to obtain the corrected BIM model; The newly added building information is aligned with the measured coordinates corresponding to the corrected BIM model. Then, the aligned new building information is input into the corrected BIM model to construct the site and obtain the construction data of the new building information.

Images