Construction deviation high-precision detection system and method based on three-dimensional laser scanning

Abstract

The invention discloses a construction deviation high-precision detection system and method based on three-dimensional laser scanning, and the system comprises a point cloud collection module, a data preprocessing module, a noisy point elimination module, a deviation analysis module, and a report generation module. The method comprises the following steps: carrying out noisy point removal on point cloud data according to regions and components, registering a processed lightweight point cloud model with a BIM model, calculating a deviation value from a point cloud to a design surface by adopting a distance in a normal direction, enabling a result to directly correspond to construction acceptance indexes such as perpendicularity and flatness, and automatically generating a visual comprehensive analysis report through a report generation module. And the coordinate of the maximum deviation point is output for structural safety recheck, so that high-precision, traceable and lightweight intelligent evaluation of the whole-field construction quality is realized, the problems of multiple interferents, contrast distortion, extensive analysis and the like in a complex construction field are effectively solved, and the method has a relatively high popularization value.

Description

Technical Field

[0001] This invention relates to the field of digital inspection technology for building engineering, and in particular to a high-precision detection system and method for construction deviation based on three-dimensional laser scanning. Background Technology

[0002] In the process of building construction quality management, detecting geometric deviations between the construction results and the design model is a core aspect of ensuring structural safety and functionality. Traditional detection methods, such as total station measurement and straightedge inspection, suffer from drawbacks such as low sampling rates, low efficiency, and strong subjectivity, making them unsuitable for the refined management requirements of modern large-scale projects. In recent years, 3D laser scanning technology has been increasingly applied to construction quality inspection due to its advantages such as non-contact operation, full-field acquisition, and high-density point cloud capabilities. However, it still faces many challenges in practical applications:

[0003] Severe on-site interference: Construction sites commonly contain non-structural objects such as steel bar stacks, formwork frames, scaffolding, and unremoved post-cast strip supports, resulting in a large amount of noise in the scanned point cloud, which directly affects the authenticity of the point cloud-BIM comparison results;

[0004] The data redundancy is enormous: a single project often generates hundreds of millions or even billions of points. Traditional denoising methods (such as statistical filtering and radius filtering) cannot accurately identify and remove unstructured points, resulting in low processing efficiency and high resource consumption.

[0005] The comparison method is unreasonable: Existing software generally uses Euclidean distance to measure deviation, that is, the shortest straight line distance from a point to the model, but this method cannot reflect the actual indicators that are of concern in project acceptance, such as "whether the wall is vertical" and "whether the ceiling is flat".

[0006] Lack of component-level differential analysis: Beams, slabs, and columns have different allowable deviations in construction specifications (e.g., columns ±10mm, slabs ±5mm), which need to be classified and processed by the design unit for targeted review. However, the existing process is mostly a whole comparison and lacks semantic breakdown capability.

[0007] Therefore, there is an urgent need for a new point cloud processing method that is designed for real construction scenarios and has the capabilities of intelligent noise reduction, component separation, and semantic deviation analysis, in order to improve the accuracy, professionalism, and practicality of the detection results. Summary of the Invention

[0008] To overcome the shortcomings of existing technologies, this invention provides a high-precision detection system and method for construction deviation based on three-dimensional laser scanning, which realizes high-precision, traceable, and lightweight intelligent assessment of the entire construction quality, effectively solving problems such as numerous interference objects, distorted comparisons, and coarse analysis in complex construction sites.

[0009] To achieve the above objectives, the present invention provides a high-precision construction deviation detection system based on three-dimensional laser scanning, comprising:

[0010] The point cloud acquisition module is used to acquire 3D laser scanning point cloud data from multiple measurement stations at the construction site;

[0011] The data preprocessing module is used to preprocess the acquired point cloud data;

[0012] The noise removal module is used to remove noise from the preprocessed point cloud data in order to clean up the detection interference items and finally export a lightweight point cloud model.

[0013] The deviation analysis module is used to calculate the distance from each point to be measured in the point cloud model to the normal direction of the BIM model, thereby obtaining the construction deviation value of each point to be measured.

[0014] The report generation module is used to analyze the construction deviations of each test point and generate a visual comprehensive analysis report from the analysis results.

[0015] Preferably, it also includes a result feedback module, which is used to extract and output the spatial coordinates of the maximum deviation point based on the visualization comprehensive analysis report, so as to conduct structural safety verification.

[0016] A high-precision detection method for construction deviations based on three-dimensional laser scanning is disclosed. This method utilizes a three-dimensional laser scanning-based high-precision detection system to achieve high-precision detection of construction deviations. The detection method includes the following steps:

[0017] The point cloud acquisition module acquires 3D laser scanning point cloud data from multiple measurement stations at the construction site.

[0018] The acquired point cloud data is preprocessed using a data preprocessing module. The preprocessing methods include:

[0019] The construction section is divided into different detection areas, and the point cloud data is spatially divided according to the different detection areas to eliminate interference from non-target areas;

[0020] The building structure point cloud data is semantically split into multiple independent subsets corresponding to different component types;

[0021] The noise removal module removes noise from independent subsets of each component to clean up detection interference items.

[0022] The cleaned point cloud data of each component is re-integrated and exported after forming a lightweight point cloud model that only contains structural point clouds.

[0023] The exported lightweight point cloud model is spatially registered with the architectural design BIM model, and the deviation analysis module is used to calculate the distance from each point to be measured in the point cloud model to the normal direction of the BIM model, thereby obtaining the construction deviation value of each point to be measured.

[0024] According to the construction specifications for different component types, corresponding dynamic tolerance thresholds are set. The construction deviation values ​​of each test point are compared and analyzed with the corresponding dynamic tolerance thresholds. The analysis results are then used to generate a visual comprehensive analysis report through the report generation module.

[0025] Preferably, before dividing the testing areas according to the construction, the acquired point cloud data is imported into point cloud data processing software to complete automatic registration and step sampling.

[0026] Preferably, when the noise removal module removes noise from independent subsets of each component to clean up detection interference, the removal method includes:

[0027] For column point clouds, determine the bottom elevation of the highest beam in the current floor, retain only the column point clouds within the range from the floor slab of the current floor to the bottom elevation of that beam, and clear the non-structural point clouds within this range;

[0028] For beam and slab point clouds, use the bottom elevation of the upper floor slab as the dividing interface to clean up the non-structural point clouds within the range from the bottom of the beam to that bottom elevation.

[0029] Preferably, when calculating the distance from each test point in the point cloud model to the normal direction of the BIM model through the deviation analysis module, thereby obtaining the construction deviation value of each test point, the calculation method includes:

[0030] Let P be any point to be measured in the point cloud model;

[0031] Find the projection point Q of point P onto the surface closest to the BIM model;

[0032] The distance d from point P to the surface is calculated using Formula 1 and used as the construction deviation value for point P.

[0033] (Formula 1)

[0034] in, Let P be the coordinates of the point to be measured. Let Q be the coordinates of the projection point. Let be the unit normal vector of the surface at the projection point Q.

[0035] Preferably, when comparing and analyzing the construction deviation values ​​of each test point with the corresponding dynamic tolerance threshold, the allowable deviation for columns is set to ±10mm, the allowable deviation for beams is set to ±8mm, and the allowable deviation for slabs is set to ±5mm.

[0036] Preferably, when generating a visualized comprehensive analysis report from the analysis results through the report generation module, the content of the visualized comprehensive analysis report includes: a three-dimensional color patch map, a detailed cross-sectional view, a list of non-compliant components, a pass rate statistics table, and the spatial coordinates of the maximum deviation point.

[0037] Preferably, after generating the visual comprehensive analysis report, the report is accessed through a device terminal, and users can jump to the corresponding three-dimensional location by clicking on the colored area, thereby realizing a closed loop of quality management where problems can be located and responsibilities can be traced.

[0038] By adopting the above technical solution, the present invention has the following beneficial effects:

[0039] 1) Precise and efficient noise reduction: By using the strategy of "vertical cutting based on the maximum beam height", typical interference objects such as frames and templates are accurately removed, with a noise reduction rate of over 85% without damaging the structural features.

[0040] 2) The analysis is more engineering-significant: deviation in the normal direction directly maps to construction and acceptance indicators, enhancing the professional interpretability of the test results.

[0041] 3) Supports differentiated evaluation: Component-level splitting allows tolerances to be set separately for beams, slabs and columns, which meets national standard requirements.

[0042] 4) Achieve full-process automation: From point cloud input to report output, no manual intervention is required, significantly shortening the detection cycle.

[0043] 5) Balancing lightweight design and high performance: Maintains the integrity of key features under extremely large data compression, making it suitable for large-scale project deployments.

[0044] 6) Supports structural safety verification: Provides clear extreme point coordinates and context information to help design units respond quickly. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0046] Figure 1 This is a flowchart illustrating the steps of a high-precision construction deviation detection method based on three-dimensional laser scanning in an embodiment of the present invention. Detailed Implementation

[0047] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings. It should be noted that these descriptions are for the purpose of aiding understanding the present invention, but do not constitute a limitation thereof. Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0048] Please see Figure 1 As shown in the figure, this embodiment of the invention provides a high-precision construction deviation detection system based on three-dimensional laser scanning, including a point cloud acquisition module, a data preprocessing module, a noise removal module, a deviation analysis module, and a report generation module. The point cloud acquisition module acquires three-dimensional laser scanning point cloud data from multiple measurement stations at the construction site. The data preprocessing module preprocesses the acquired point cloud data. The noise removal module removes noise from the preprocessed point cloud data to eliminate detection interference and ultimately export a lightweight point cloud model. The deviation analysis module calculates the distance from each point to be measured in the point cloud model to the normal direction of the BIM model, thereby obtaining the construction deviation value of each point. The report generation module analyzes the construction deviation of each point and generates a visualized comprehensive analysis report from the analysis results.

[0049] Furthermore, in this embodiment, the high-precision construction deviation detection system based on three-dimensional laser scanning also includes a result feedback module. The result feedback module is used to extract and output the spatial coordinates of the maximum deviation point according to the visualization comprehensive analysis report, so as to conduct structural safety verification.

[0050] This invention also discloses a high-precision detection method for construction deviations based on three-dimensional laser scanning. The method achieves high-precision detection of construction deviations using the aforementioned three-dimensional laser scanning-based high-precision detection system. The detection method includes the following steps:

[0051] The point cloud acquisition module acquires 3D laser scanning point cloud data from multiple measurement stations at the construction site. The acquired point cloud data is in FLS format.

[0052] The acquired point cloud data is preprocessed using a data preprocessing module. The preprocessing methods include:

[0053] The construction section is divided into different detection areas, and the point cloud data is spatially divided according to the different detection areas to eliminate interference from non-target areas;

[0054] The building structure point cloud data is semantically split into multiple independent subsets corresponding to different component types;

[0055] The noise removal module removes noise from independent subsets of each component to clean up detection interference items.

[0056] The cleaned point cloud data of each component is re-integrated and exported after forming a lightweight point cloud model that only contains structural point clouds.

[0057] The exported lightweight point cloud model is spatially registered with the architectural design BIM model, and the deviation analysis module is used to calculate the distance from each point to be measured in the point cloud model to the normal direction of the BIM model, thereby obtaining the construction deviation value of each point to be measured.

[0058] According to the construction specifications for different component types, corresponding dynamic tolerance thresholds are set. The construction deviation values ​​of each test point are compared and analyzed with the corresponding dynamic tolerance thresholds. The analysis results are then used to generate a visual comprehensive analysis report through the report generation module.

[0059] Furthermore, before dividing the detection areas according to the construction, the acquired point cloud data is imported into point cloud data processing software (Trimble RealWorks software in this embodiment) to complete automatic registration and step-size sampling. Step-size sampling can reduce the density of the initial data. Preferably, in this embodiment, the building structure point cloud data is semantically divided into three independent subsets: beams, slabs, and columns according to different component types. When the noise removal module removes noise from each independent subset of the component to clean up detection interference items, the removal method includes:

[0060] For column point clouds, analyze the height of all beams on the current floor to determine the bottom elevation of the highest beam on the current floor. Only retain the vertically continuous column point cloud within the range from the floor slab of the current floor to the bottom elevation of the beam (i.e., the cleanup range is [floor slab elevation of the current floor, bottom elevation of the highest beam]). Remove non-structural point clouds within this range, specifically including: ground debris, mobile equipment, scaffold legs, and other lateral interference points. This operation can ensure that column detection is not affected by the accumulation of materials on the ground floor.

[0061] For beam and slab point clouds, use the bottom elevation of the upper floor slab as the dividing interface to clean up the non-structural point clouds within the range from the bottom of the beam to that bottom elevation (i.e., focus on processing the "beam bottom to slab bottom" interval). Specifically, this includes the formwork system and support frame in the post-cast strip area. Since such structures cannot be removed before the structure is capped, they will inevitably exist in the scan data. If they are not removed, the pass rate will be seriously low.

[0062] It should be noted that the cleaned point cloud data of each component is re-integrated and exported into the standard LAS format. The point cloud and BIM model are then accurately registered in Cyclone_3DR using the known target sphere coordinates (such as A(X=6098.380, Y=8181.908, Z=2079.988)).

[0063] The deviation analysis module is integrated into the Cyclone_3DR or CloudWorx platform, supporting data interaction with mainstream BIM software such as Revit / Bentley. When calculating the distance from each point in the point cloud model to the normal direction of the BIM model, thereby obtaining the construction deviation value of each point, the calculation method includes:

[0064] Let P be any point to be measured in the point cloud model;

[0065] Find the projection point Q of point P onto the surface closest to the BIM model;

[0066] The distance d from point P to the surface is calculated using Formula 1 and used as the construction deviation value for point P.

[0067] (Formula 1)

[0068] in, Let P be the coordinates of the point to be measured. Let Q be the coordinates of the projection point. Let be the unit normal vector of the surface at the projection point Q.

[0069] By replacing the traditional Euclidean distance calculation method with normal direction deviation analysis, the normal direction distance can be used as a standard measure of deviation. The advantage of this method is that:

[0070] If the BIM surface is a vertical wall, then the normal is in the horizontal direction, and the deviation d represents the "verticality" error.

[0071] If the BIM surface is a horizontal floor slab, then the normal is in the vertical direction, and the deviation d represents the "flatness" or "elevation" error.

[0072] This represents a leap from "geometric differences" to "quality interpretation," making the test results more meaningful for engineering applications.

[0073] Furthermore, when comparing the construction deviation values ​​of each test point with the corresponding dynamic tolerance threshold, the allowable deviation for columns is set to ±10mm, for beams to ±8mm, and for slabs to ±5mm, which complies with the relevant provisions of the "Code for Acceptance of Construction Quality of Concrete Structures" GB50204. Preferably, when generating a visual comprehensive analysis report from the analysis results through the report generation module, the format of the visual comprehensive analysis report is PDF / HTML, which can be used for quality assessment. For rectification tracking and design review, the content of the visual comprehensive analysis report includes: a 3D color patch map (showing the degree of deviation in each area using a color spectrum), a detailed cross-sectional view (slicing along a specified path to show the deviation trend), a list of non-compliant components (listing the number, location, and maximum deviation value of all components exceeding the allowable deviation range), a pass rate statistics table (summarizing the compliance rate by component category), and the spatial coordinates of the maximum deviation point (outputting XYZ coordinates, supporting reverse lookup to the specific location on site). It should be noted that in this embodiment, after the visualization comprehensive analysis report is generated, the visualization comprehensive analysis report is accessed through the device terminal, and the corresponding three-dimensional position is jumped by clicking on the color spot area, thereby realizing a quality management closed loop where problems can be located and responsibilities can be traced. For example, the coordinates of the maximum deviation point and its contextual information (component, deviation direction, and adjacent structural status) are packaged and sent to the design institute to help determine whether it affects the bearing capacity, crack control, or seismic performance, thus realizing a construction-design collaborative closed loop.

[0074] The following example, using an airport project, further illustrates the specific implementation process of this invention:

[0075] 1) Project Overview

[0076] 1. Data import and registration: Import 6 FLS files into Trimble RealWorks, set the step sampling interval to 5mm, enable the automatic registration function, and the registration accuracy RMS < 2mm.

[0077] 2. Region division and component decomposition: Delineate the detection boundaries in the software and use semantic tags to divide the point cloud into three categories: "column", "beam" and "slab".

[0078] 3. Column Noise Removal: The highest beam height on this floor was found to be 850mm, the beam bottom elevation was 3.150m, and the footboard elevation was ±0.000m. The cleaning range was set to [0.000m, 3.150m], and non-column point clouds (such as steel pipes, timber, support frames, and edge protection) within this range were removed. After processing, the column point cloud retention rate was approximately 92%, and the interference point removal rate was approximately 96%.

[0079] 4. Beam and slab noise removal: The bottom elevation of the upper floor slab is 4.000m. Focus on processing the point cloud within the range of 3.150m to 4.000m, manually select and delete the post-cast strip formwork and U-shaped brackets. After completion, the amount of beam and slab point cloud data decreased by approximately 80%.

[0080] 5. Point cloud recombination and export: Merge the point clouds of the three types of components and export them in LAS format, reducing the total number of points to about 50% of the original data.

[0081] 6. Point Cloud-BIM Registration: Import the Revit model and match the coordinates of 4 target points in Cyclone_3DR to achieve high-precision alignment.

[0082] 7. Normal Deviation Analysis: Set the normal deviation analysis mode to calculate the normal deviation at each point. Results display:

[0083] Average column deviation: +3.2mm, maximum deviation: +12.4mm (exceeding the ±10mm limit)

[0084] Average board deviation: -2.1mm, maximum deviation: -6.8mm (exceeding the ±5mm limit)

[0085] 8. Generate a comprehensive report:

[0086] The output is a PDF report containing the following: a rainbow chart showing the overall deviation distribution; two typical cross-sectional views; a list of non-compliant components (7 in total); a pass rate statistics table (columns: 91.3%, beams: 96.7%, slabs: 88.2%); and a list of coordinates of the points with the largest deviations.

[0087] 9. Structural safety feedback: Submit the coordinates of the maximum deviation point (X=6098.380, Y=8181.908, Z=2079.988) and related screenshots to the design institute. After review and confirmation that it does not affect structural safety, it is agreed to proceed to the next process.

[0088] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and these variations still fall within the protection scope of the present invention.

Claims

1. A high-precision construction deviation detection system based on three-dimensional laser scanning, characterized in that, include: The point cloud acquisition module is used to acquire 3D laser scanning point cloud data from multiple measurement stations at the construction site; The data preprocessing module is used to preprocess the acquired point cloud data; The noise removal module is used to remove noise from the preprocessed point cloud data in order to clean up the detection interference items and finally export a lightweight point cloud model. The deviation analysis module is used to calculate the distance from each point to be measured in the point cloud model to the normal direction of the BIM model, thereby obtaining the construction deviation value of each point to be measured. The report generation module is used to analyze the construction deviations of each test point and generate a visual comprehensive analysis report from the analysis results.

2. The high-precision construction deviation detection system based on three-dimensional laser scanning as described in claim 1, characterized in that, It also includes a result feedback module, which is used to extract and output the spatial coordinates of the maximum deviation point based on the visualization comprehensive analysis report, so as to conduct structural safety verification.

3. A high-precision detection method for construction deviations based on three-dimensional laser scanning, characterized in that, The high-precision detection system for construction deviations based on three-dimensional laser scanning, as described in claim 1, achieves high-precision detection of construction deviations. The detection method includes the following steps: The point cloud acquisition module acquires 3D laser scanning point cloud data from multiple measurement stations at the construction site. The acquired point cloud data is preprocessed using a data preprocessing module. The preprocessing methods include: The construction section is divided into different detection areas, and the point cloud data is spatially divided according to the different detection areas to eliminate interference from non-target areas; The building structure point cloud data is semantically split into multiple independent subsets corresponding to different component types; The noise removal module removes noise from independent subsets of each component to clean up detection interference items. The cleaned point cloud data of each component is re-integrated and exported after forming a lightweight point cloud model that only contains structural point clouds. The exported lightweight point cloud model is spatially registered with the architectural design BIM model, and the deviation analysis module is used to calculate the distance from each point to be measured in the point cloud model to the normal direction of the BIM model, thereby obtaining the construction deviation value of each point to be measured. According to the construction specifications for different component types, corresponding dynamic tolerance thresholds are set. The construction deviation values ​​of each test point are compared and analyzed with the corresponding dynamic tolerance thresholds. The analysis results are then used to generate a visual comprehensive analysis report through the report generation module.

4. The high-precision detection method for construction deviation based on three-dimensional laser scanning as described in claim 3, characterized in that, Before dividing the testing areas according to the construction, the acquired point cloud data is imported into the point cloud data processing software to complete automatic registration and step sampling.

5. The high-precision detection method for construction deviation based on three-dimensional laser scanning as described in claim 3, characterized in that, When noise removal is performed on independent subsets of each component using the noise removal module to clean up detection interference, the removal methods include: For column point clouds, determine the bottom elevation of the highest beam in the current floor, retain only the column point clouds within the range from the floor slab of the current floor to the bottom elevation of that beam, and clear the non-structural point clouds within this range; For beam and slab point clouds, use the bottom elevation of the upper floor slab as the dividing interface to clean up the non-structural point clouds within the range from the bottom of the beam to that bottom elevation.

6. The high-precision detection method for construction deviation based on three-dimensional laser scanning as described in claim 3, characterized in that, When calculating the distance from each point to be measured in the point cloud model to the normal direction of the BIM model through the deviation analysis module, and thus obtaining the construction deviation value of each point to be measured, the calculation method includes: Let P be any point to be measured in the point cloud model; Find the projection point Q of point P onto the surface closest to the BIM model; The distance d from point P to the surface is calculated using Formula 1 and used as the construction deviation value for point P. (Formula 1) in, Let P be the coordinates of the point to be measured. Let Q be the coordinates of the projection point. Let be the unit normal vector of the surface at the projection point Q.

7. The high-precision detection method for construction deviation based on three-dimensional laser scanning as described in claim 3, characterized in that, When comparing and analyzing the construction deviation values ​​of each test point with the corresponding dynamic tolerance threshold, the allowable deviation for columns is set to ±10mm, the allowable deviation for beams is set to ±8mm, and the allowable deviation for slabs is set to ±5mm.

8. The high-precision detection method for construction deviation based on three-dimensional laser scanning as described in claim 3, characterized in that, When generating a visual comprehensive analysis report from the analysis results through the report generation module, the content of the visual comprehensive analysis report includes: a three-dimensional color patch map, a detailed cross-sectional view, a list of non-compliant components, a pass rate statistics table, and the spatial coordinates of the maximum deviation point.

9. The high-precision detection method for construction deviation based on three-dimensional laser scanning as described in claim 8, characterized in that, After generating the visual comprehensive analysis report, the report is accessed through a device terminal, and users can jump to the corresponding three-dimensional location by clicking on the colored area, thereby realizing a closed loop of quality management where problems can be located and responsibilities can be traced.

Images