Construction project engineering management audit doubtful point extraction method based on three-dimensional point cloud

Through the automated comparison of three-dimensional point cloud and BIM model and semantic segmentation model detection, the problems of low audit efficiency and insufficient accuracy in the existing technology are solved, efficient and accurate extraction of suspicious points in engineering management are achieved, and construction quality and progress are improved Comprehensive and in-depth evaluation.

CN119067616AActive Publication Date: 2024-12-03QINGDAO INST OF SURVEYING & MAPPING SURVEY +1

Patent Information

Application Number
CN202411563276.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-05
Publication Date
2024-12-03
Estimated Expiration
2044-11-05

AI Technical Summary

Technical Problem

There are problems of low efficiency and insufficient accuracy in the existing construction project engineering management audit methods, especially the time delay and error caused by accuracy loss and manual intervention when comparing and analyzing point cloud data with BIM models.

Method used

Adopt the three-dimensional point cloud-based engineering management audit doubt point extraction method, through data preparation, BIM model geographical location definition, spatial reference transformation, semantic segmentation model training and automated detection, to realize automated comparison and doubt point detection of point cloud data and BIM model, reducing Manual intervention and improve accuracy.

Benefits of technology

It improves audit efficiency and accuracy, reduces time and cost, can comprehensively evaluate construction quality and progress from multiple dimensions, detects dimensional deviations, material deviations and unfinished parts during the construction process, and improves the comprehensiveness and in-depthness of the audit.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067616A_ABST
    Figure CN119067616A_ABST
Patent Text Reader

Abstract

The invention relates to the field of engineering auditing, in particular to a construction project engineering management auditing doubtful point extraction method based on three-dimensional point clouds, which comprises the following steps: step 1, constructing and forming an auditing data resource pool based on a BIM model as a main body; 2, placing the two types of non-homologous data in the same coordinate system; 3, associating geometric data of the BIM model with project management information; 4, pre-training the laser point cloud semantic segmentation model, and accurately identifying key components in the actually measured point cloud in the auditing stage; and step 5, processing the point cloud data, identifying key engineering components, performing geographic space matching and fusion on identification results and BIM model component point cloud, calculating and comparing geometric feature differences of the key components, and completing doubtful point detection. By introducing an advanced three-dimensional point cloud and machine learning technology, an efficient, accurate and intelligent construction project engineering management auditing doubtful point extraction method is provided, and powerful technical support is provided for engineering management and auditing work.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of engineering management audit, and in particular to a method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud. Background Art

[0002] Construction project engineering management audit refers to the audit of the quality, safety and progress of engineering construction by auditors in accordance with relevant laws, regulations and industry norms, as well as relevant design contract documents of engineering construction, during the project construction process, to promote the smooth realization of project construction goals. The audit content mainly includes reviewing the implementation of the construction project progress management plan and reviewing the quality acceptance of the construction process. In modern construction projects, with the continuous increase in the scale and complexity of projects, audit evidence collection faces greater challenges. At present, many regions have begun to explore the use of laser point cloud and 3D modeling technology to carry out audit work. By constructing a 3D model through the acquired point cloud data, and comparing and analyzing it with the real-life model and BIM model, potential audit doubts can be preliminarily screened out. This technical method not only achieves comprehensive coverage of project engineering, greatly reduces the workload of field verification, improves work efficiency, but also significantly reduces measurement errors, thereby effectively ensuring the quality of audits.

[0003] Although the above technical methods have certain advantages in audit evidence collection, there are still several problems. For example, in the process of building a three-dimensional model based on point cloud data, there will be a loss of accuracy, which may cause accuracy differences when comparing and analyzing with the BIM model later. In addition, the modeling process usually requires a lot of time and manpower, resulting in time delays in audit evidence collection. Furthermore, the current combined application of data mostly stays at the stage of visual display and simple comparative analysis, requiring a lot of manual intervention, not only with a low degree of automation, but also may introduce human errors. To date, there is still a lack of methods that combine point cloud technology, BIM technology and intelligent algorithms to achieve automated and intelligent audit evidence collection. Summary of the invention

[0004] In view of the problems of low efficiency and insufficient accuracy in the existing construction project engineering management audit evidence collection methods, the purpose of the present invention is to provide a method for extracting audit doubts based on three-dimensional point clouds, which is directly based on the comparison of the collected point cloud data with the BIM model analysis point cloud, reduces the accuracy loss and time consumption in the three-dimensional modeling link, and uses machine learning methods to automatically extract and identify audit doubts in engineering management, thereby improving audit efficiency and accuracy.

[0005] To achieve the above purpose, the technical solution adopted by the present invention is: a method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud, comprising the following steps: Step 1: Audit data preparation, data collection, organization and storage, building an audit data resource pool based on the BIM model, and accurately constructing a three-dimensional portrait of different time stages during the project implementation; Step 2: Define the geographic location of the BIM model and convert the spatial reference. Define the geographic location of the BIM model and convert it to the laser scanning reference coordinate system, placing the two types of non-homologous data in the same coordinate system. Step 3: BIM model parses component point cloud and associated project management attributes, parses the geometric data of the BIM model into a three-dimensional point cloud, and associates it with project management information; Step 4: Train the laser point cloud semantic segmentation model. Pre-train the laser point cloud semantic segmentation model based on machine learning methods to accurately identify key components in the measured point cloud during the audit phase. Step 5: Perform automatic detection of suspicious points based on the measured point cloud and the BIM component point cloud. Use the semantic segmentation model obtained in step 4 to process the point cloud data of each on-site measurement, automatically identify the key engineering components in the construction project, and perform geospatial matching and fusion of the identification results and the BIM model component point cloud. Use automated calculation methods to compare the geometric feature differences of the key components in the two source point cloud data to complete the suspicious point detection.

[0006] The above-mentioned method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud also includes step 6: doubtful point elimination, labeling and visual output, and human-computer interaction verification and elimination of doubtful points detected automatically. By checking relevant evidence or records, it is determined whether the doubtful points are valid. If it is found that the doubtful points are caused by design changes, data entry errors or other normal business activities, these doubtful points will be eliminated.

[0007] The above-mentioned construction project engineering management audit doubt point extraction method based on three-dimensional point cloud, step 1 comprises: Step 1-1: Obtain relevant content of the project through business systems, regulatory platforms and Internet channels, conduct data governance, record key time nodes of the project, and convert the relevant content and key time node information into structured data for storage as project feature input; Step 1-2: Use the 3D model built using BIM software during the project design phase to add component material information based on the design instructions and construction specifications; Step 1-3: Collect laser point cloud data from the early stage of the project as the training data source for the semantic segmentation model, record and save the results of multiple laser point cloud phases, and form a complete data chain.

[0008] In the above-mentioned method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud, step 2 comprises: Step 2-1: Clarify the projection coordinate system used in the project design and the geographic coordinate system to be converted; Step 2-2: Set the plane coordinate information of the model data through BIM software to determine its latitude, longitude and altitude; Step 2-3: Obtain the angle between the project north and true north through drawing measurement, which is used as the angle between the BIM model and true north; Step 2-4: Use conversion tools or conversion algorithms to convert the BIM model from planimetric coordinates to geospatial coordinates.

[0009] In the above-mentioned method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud, step 3 comprises: Step 3-1: Filter the key component categories involved in the audit and detection based on semantic information, extract the corresponding BIM model components and convert them into triangular surface models, perform topological structure correction and data compression after conversion, reduce the number of triangles, and reduce the complexity of the model; Step 3-2: Determine the sampling point density, capture the curvature changes of the geometric surface, and realize the three-dimensional point cloud expression of the BIM model components through the closed contour point cloud filling method; Step 3-3: Assign sampling point attributes, associate the sampling points with project management information, and export them to a standard point cloud file format.

[0010] In the above-mentioned method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud, in step 4, pre-training includes data set preparation and model training, and the specific steps are as follows: Step 4-1: Manually annotate the point cloud data collected in the early stage, generate point cloud samples covering important categories of components and their corresponding semantic labels, and integrate them into the existing 3D public dataset to form a high-quality training dataset covering a variety of scenarios; Step 4-2: Consider the accuracy of local feature extraction and the efficiency of global semantic segmentation, and select an appropriate training model framework; Step 4-3: Preprocess the point cloud data and divide the data set into training set, validation set and test set; Step 4-4: Define the loss function and optimization strategy, optimize the model step by step, adjust the hyperparameters using the validation set, improve the generalization ability and robustness of the BIM model, and evaluate the BIM model using the test set.

[0011] In the above-mentioned method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud, in step 5, the detection content includes: detecting position and / or size deviation, detecting material deviation, and detecting unfinished parts.

[0012] The above-mentioned construction project engineering management audit suspicious point extraction method based on three-dimensional point cloud, the detection position and / or size deviation is determined by calculating the symmetrical distance between two clusters of point clouds of the same component in the adjacent space to determine whether there is a position deviation. When the distance exceeds the set threshold, it can be determined that the component position has deviated. The specific method is: Step a: Assume two clusters of point clouds, where point cloud A contains the point set , point cloud B contains the point set , for each point in point cloud A , find the nearest point in point cloud B , and calculate its distance. Similarly, for each point in point cloud B , find the nearest point in point cloud A ; Step b: The average closest point distance from point cloud A to point cloud B: , where m is a constant; Step c: The average closest point distance from point cloud B to point cloud A: , where n is a constant; Step d: Symmetric distance calculation: .

[0013] The above-mentioned construction project engineering management audit suspicious point extraction method based on three-dimensional point cloud, the detection of material deviation includes extracting the color features of each component in the measured point cloud, and searching for the corresponding BIM model component point cloud based on its spatial position to obtain the corresponding material semantic information, and comparing it with the point cloud data of the same material label in the training data set. The specific steps are as follows: Step e: discretize the color value into multiple color intervals, calculate the number or frequency of point clouds in each color interval, and obtain a color histogram as a description of the color feature; Step f: Use the distance metric formula , compare the color histogram differences of the two groups of point clouds and quantify the material deviation, where and Respectively represent the histogram values ​​of point cloud A and point cloud B in the color range.

[0014] In the above-mentioned method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud, the unfinished part of the inspection includes: Step g: Filter the components that should have been built at the current time in the BIM model point cloud by combining the construction period time semantics, and search for the measured point cloud that matches it based on the spatial position; Step h: quantify the coverage of the point clouds by calculating the intersection-and-union ratio between the two, voxelize the two point clouds respectively, assign each point to the corresponding voxel, and calculate the intersection and union of the two sets of voxel sets. The overlap is expressed as: , when the calculation result is lower than a certain threshold, it can be determined that the component is not completed.

[0015] The invention discloses a method for extracting doubtful points in the audit of construction project engineering management based on three-dimensional point cloud, which has the following beneficial effects: by constructing three-dimensional portraits, scene reproduction and data tracing at different times and spatial reference conversion are realized, which lays the foundation for realizing accurate fusion and comparative analysis of the two in spatial dimensions during the audit stage. Through automated anomaly detection, the audit efficiency is greatly improved, and the time and cost of manual intervention are reduced; by using the method combining three-dimensional point cloud data and BIM model, high-precision geometric information can be provided, and by introducing machine learning technology, historical data can be trained and learned to establish an intelligent anomaly detection model. The model can continuously optimize itself, and with the increase of data volume and diversity, the detection accuracy and reliability will also be continuously improved, thereby improving the accuracy of audit evidence collection; and the invention extracts doubtful points from multiple dimensions, which can not only detect the size deviation and material deviation in the construction process, but also identify the unfinished parts, comprehensively evaluate the construction quality and progress from multiple dimensions, and improve the comprehensiveness and depth of the audit; by realizing three-dimensional visualization of the evidence collection results, the doubtful point records generated by the extraction are intuitive and easy to understand, that is, they contain the detected abnormal conditions and specific locations, and can be combined with graphic display, which is convenient for auditors to quickly understand and judge.

[0016] By introducing advanced three-dimensional point cloud and machine learning technology, the present invention provides an efficient, accurate and intelligent method for extracting doubtful points in construction project engineering management auditing, which provides strong technical support for engineering management and auditing work and has significant application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention is a schematic diagram of the process of extracting doubtful points in the construction project engineering management audit based on three-dimensional point cloud. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is described below in conjunction with specific implementation methods and drawings.

[0019] Figure 1 The paper fully describes the process of extracting doubtful points in construction project engineering management audit based on 3D point cloud. Figure 1 , the extraction method of the present invention is as follows.

[0020] Example 1 A method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud, comprising the following steps: Step 1: Audit data preparation, carry out data collection, organization and storage, build an audit data resource pool based on the BIM model, and accurately construct a three-dimensional portrait of different time stages during the project implementation.

[0021] Step 1-1: Obtain relevant content of the project through business systems, regulatory platforms and Internet channels, conduct data governance, record key time nodes of the project, and convert the information of relevant content and key time nodes into structured data for storage as project feature input.

[0022] Step 1-2: Use the 3D model built using BIM software during the project design phase to add component material information based on the design instructions and construction specifications.

[0023] Step 1-3: Collect laser point cloud data from the early stage of the project as the training data source for the semantic segmentation model, record and save the results of multiple laser point cloud phases, and form a complete data chain.

[0024] Step 2: Define the geographic location of the BIM model and convert the spatial reference. Define the geographic location of the BIM model and convert it to the laser scanning reference coordinate system, placing the two types of non-homologous data in the same coordinate system.

[0025] Step 2-1: Clarify the projection coordinate system used in the project design and the geographic coordinate system to be converted.

[0026] Step 2-2: Set the plane coordinate information of the model data through BIM software to determine its latitude, longitude and altitude.

[0027] Step 2-3: Obtain the angle between the project north and true north through measurement on the drawings, which will be used as the angle between the BIM model and true north.

[0028] Step 2-4: Use conversion tools or conversion algorithms to convert the BIM model from planimetric coordinates to geospatial coordinates.

[0029] Step 3: BIM model parses component point cloud and associated project management attributes, parses the geometric data of the BIM model into a three-dimensional point cloud, and associates it with project management information.

[0030] Step 3-1: Filter the key component categories involved in the audit and detection based on semantic information, extract the corresponding BIM model components and convert them into triangular surface models. After conversion, perform topological structure correction and data compression to reduce the number of triangles and reduce model complexity.

[0031] Step 3-2: Determine the sampling point density, capture the curvature changes of the geometric surface, and realize the three-dimensional point cloud expression of BIM model components through closed contour point cloud filling.

[0032] Step 3-3: Assign sampling point attributes, associate the sampling points with project management information, and export them to a standard point cloud file format.

[0033] Step 4: Train the laser point cloud semantic segmentation model. Pre-train the laser point cloud semantic segmentation model based on machine learning methods to accurately identify key components in the measured point cloud during the audit phase.

[0034] Pre-training includes data set preparation and model training. The specific steps are: Step 4-1: Manually annotate the point cloud data collected in the early stage to generate point cloud samples covering important categories of components and their corresponding semantic labels, and integrate them into the existing 3D public dataset to form a high-quality training dataset covering a variety of scenarios.

[0035] Step 4-2: Consider the accuracy of local feature extraction and the efficiency of global semantic segmentation, and select an appropriate training model framework.

[0036] Step 4-3: Preprocess the point cloud data and divide the data set into training set, validation set and test set.

[0037] Step 4-4: Define the loss function and optimization strategy, optimize the model step by step, adjust the hyperparameters using the validation set, improve the generalization ability and robustness of the BIM model, and evaluate the BIM model using the test set.

[0038] Step 5: Perform automatic detection of suspicious points based on the measured point cloud and the BIM component point cloud. Use the semantic segmentation model obtained in step 4 to process the point cloud data of each on-site measurement, automatically identify the key engineering components in the construction project, and perform geospatial matching and fusion of the identification results and the BIM model component point cloud. Use automated calculation methods to compare the geometric feature differences of the key components in the two source point cloud data to complete the suspicious point detection.

[0039] The inspection contents include: inspection of position and / or size deviation, inspection of material deviation, and inspection of unfinished parts.

[0040] Step 6: Eliminate doubts, mark and visualize output. Perform human-computer interaction verification and elimination on the doubts detected automatically. By checking relevant evidence or records, determine whether the doubts are valid. If it is found that the doubts are caused by design changes, data entry errors or other normal business activities, these doubts will be eliminated.

[0041] Example 2 By comparing the differences between the current 3D scene of the construction site and the pre-construction design model, audit doubts are discovered and extracted. The current 3D scene of the construction site is point cloud data obtained through laser scanning technology, and the design model is a BIM model.

[0042] 1. Preparation of audit data.

[0043] Carry out data collection, organization and storage work, and build an audit data resource pool with three-dimensional models as the main body.

[0044] First, obtain the project's contract, survey report, design drawings, construction plan and other contents through business systems, regulatory platforms and the Internet, conduct data governance, record the key time nodes of the project (such as start date, completion date, and phased acceptance date), and convert this information into structured data for storage as project feature input; second, prepare the project design BIM model as a basis for audit evidence collection. The three-dimensional model that has been built using BIM software during the project design phase can be directly used, or it can be re-modeled based on the design drawings, and the material information of the components can be supplemented according to the design instructions and construction specifications. Third, collect laser point cloud data from the early stage of the project as a training data source for the semantic segmentation model to improve the accuracy of model training and overcome the limitations of existing public three-dimensional data sets.

[0045] At the same time, the results of multiple laser point clouds are recorded and saved to form a complete data chain, accurately constructing a "three-dimensional portrait" of different time stages during the implementation of the project, based on which scene reconstruction and data traceability at different times can be achieved.

[0046] 2. BIM model geographic location definition and spatial reference conversion.

[0047] Define the geographic spatial position of the BIM model and convert it to the laser scanning reference coordinate system, place the two types of non-homologous data in the same coordinate system, ensure the consistency of the spatial position between the data, and lay the foundation for the precise integration and comparative analysis of the two in the spatial dimension during the audit stage.

[0048] First, clarify the projection coordinate system used in the project design and the geographic coordinate system to be converted. Then set the plane coordinate information of the model data through the BIM software to determine its latitude, longitude and altitude. This can be achieved by setting the model project base point or specifying shared coordinates. The specific operation is to select a certain point on the model with known drawing coordinates and set the actual coordinate value of the point, so as to reposition the model in the shared coordinate system and move it relative to the global positioning link. In addition, it is necessary to set the angle between the model and the true north. The angle value is the angle between the project north and the true north, which can be obtained through drawing measurement. Subsequently, use the conversion tool or conversion algorithm to convert the model from the plane coordinates to the geospatial coordinates, and import the converted model into the geographic information system (GIS) for verification to ensure the accuracy and consistency of the conversion.

[0049] 3. BIM model analysis component point cloud and related project management attributes.

[0050] BIM models contain rich geometric structures and semantic information, but their geometric expressions are significantly different from laser point cloud data. Therefore, in order to achieve efficient and automated matching and computational analysis between BIM models and laser point cloud data, it is necessary to convert the BIM geometric data format and parse it into a three-dimensional point cloud while retaining key attribute information such as component category and material. In addition, the point cloud data generated by the analysis will also be associated with project management information such as the construction period and time nodes extracted from the project contract and construction plan.

[0051] The specific process is as follows: 1) According to the semantic information, the key component categories involved in the audit and detection, such as walls, floors, doors, windows, etc., are selected, and the corresponding BIM component models are extracted and converted into triangular surface models. After the conversion, the topological structure is corrected and the data is compressed to reduce the number of triangles and the complexity of the model.

[0052] 2) Determine the sampling point density and sample evenly within each triangular face. For curved surfaces, sample along the normal direction to accurately capture the curvature changes on the geometric surface, thereby achieving three-dimensional point cloud expression of BIM components through closed contour point cloud filling.

[0053] 3) Assign sampling point attributes, such as coordinates, normal vectors, etc., as well as extended custom attributes such as material and duration, to achieve association with project management information and export it to a standard point cloud file format.

[0054] 4. Train the laser point cloud semantic segmentation model.

[0055] The laser point cloud semantic segmentation model is pre-trained based on machine learning methods to achieve accurate identification of key components in the measured point cloud during the audit phase. The training process mainly includes two parts: data set preparation and model training.

[0056] First, the point cloud data collected in the early stage are manually annotated to generate point cloud samples covering important categories of components and their corresponding semantic labels, which are then integrated into the existing three-dimensional public dataset to form a high-quality training dataset covering a variety of scenarios to meet the point cloud segmentation needs in complex indoor architectural environments. Then, the appropriate training model framework is selected by comprehensively considering the accuracy of local feature extraction and the efficiency of global semantic segmentation.

[0057] Before model training, the point cloud data is preprocessed by normalization, downsampling, and filtering, and the data set is divided into training set, validation set, and test set. During the model training process, the model is gradually iteratively optimized by defining appropriate loss functions and optimization strategies, and the hyperparameters are adjusted using the validation set to improve the generalization and robustness of the model. Finally, the model is evaluated using the test set, and the actual performance of the model is verified based on the segmentation accuracy index to ensure that it meets the expected accuracy requirements.

[0058] 5. Automatically detect suspicious points based on measured point cloud and BIM component point cloud.

[0059] During the audit phase, the semantic segmentation model trained in the previous steps is used to process the point cloud data of each on-site measurement, automatically identify the key engineering components in the construction project, and then perform geospatial matching and fusion with the BIM component point cloud. The geometric feature differences of the key components in the point cloud data from the two sources are compared through automated calculations to complete suspicious point detection.

[0060] The specific testing content covers the following three aspects: ① Position / size deviation: By calculating the symmetrical distance between two clusters of point clouds of the same component in the adjacent space, it is determined whether there is a position deviation. When the distance exceeds the set threshold, it can be determined that the component position has deviated. Assume two clusters of point clouds, where point cloud A contains point set , point cloud B contains the point set , for each point in point cloud A , find the nearest point in point cloud B , and calculate its distance. Similarly, for each point in point cloud B , find the nearest point in point cloud A , and calculate their distance.

[0061] The calculation process is as follows: The average closest point distance from point cloud A to point cloud B: , The average closest point distance from point cloud B to point cloud A: , Symmetric distance calculation: .

[0062] ② Material deviation: Extract the color features of each component in the measured point cloud, and find the corresponding BIM component point cloud based on its spatial position to obtain the corresponding material semantic information. Then, compare it with the point cloud data with the same material label in the training data set. The specific method is to discretize the color value into multiple color intervals, calculate the number or frequency of point clouds in each color interval, and obtain the color histogram as a description of the color feature. The color histogram difference between the two groups of point clouds is compared by the distance measurement formula to quantify the material deviation.

[0063] The calculation formula is as follows: ,in, and Respectively represent point cloud A and point cloud B in the color range The histogram values ​​within .

[0064] ③ Detection of unfinished parts: Combine the semantics of construction period to filter the components that should have been completed at the current time in the BIM point cloud, and find the measured point cloud that matches it based on the spatial position. The point cloud coverage is quantified by calculating the intersection and union ratio between the two. When the calculation result is lower than a certain threshold, it can be determined that the component is unfinished. The specific calculation steps are as follows: voxelize the two point clouds separately, assign each point to the corresponding voxel, and calculate the intersection and union of the two sets of voxel sets.

[0065] The overlap is expressed as: .

[0066] 6. Eliminate doubts, mark and output visually.

[0067] Human-computer interaction verification and elimination are carried out for the suspicious points detected automatically. By checking the relevant evidence or records, it is determined whether the suspicious points are valid. If the suspicious points are found to be caused by design changes, data entry errors or other normal business activities, these suspicious points will be eliminated. Three-dimensional visual annotation is carried out for the verified suspicious points, and corresponding indicators are estimated through GIS functions such as spatial measurement, such as measuring the length, area, volume, etc. of each structure in the construction project, intuitively showing the spatial location and abnormal characteristics of each suspicious point, and enhancing the transparency and interpretability of the results. Finally, the suspicious point records are batch-output by project to support the follow-up work of the audit.

[0068] The above embodiments are only for illustrating the inventive concept and features of the present invention, and their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made based on the essence of the content of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud, characterized in that: The following steps are involved: Step 1: Audit data preparation, data collection, organization and storage, building an audit data resource pool based on the BIM model, and accurately constructing a three-dimensional portrait of different time stages during the project implementation; Step 2: Define the geographic location of the BIM model and convert the spatial reference. Define the geographic location of the BIM model and convert it to the laser scanning reference coordinate system, placing the two types of non-homologous data in the same coordinate system. Step 3: BIM model parses component point cloud and associated project management attributes, parses the geometric data of the BIM model into a three-dimensional point cloud, and associates it with project management information; Step 4: Train the laser point cloud semantic segmentation model. Pre-train the laser point cloud semantic segmentation model based on machine learning methods to accurately identify key components in the measured point cloud during the audit phase. Step 5: Perform automatic detection of suspicious points based on the measured point cloud and the BIM component point cloud. Use the semantic segmentation model obtained in step 4 to process the point cloud data of each on-site measurement, automatically identify the key engineering components in the construction project, and perform geospatial matching and fusion of the identification results and the BIM model component point cloud. Use automated calculation methods to compare the geometric feature differences of the key components in the two source point cloud data to complete the suspicious point detection.

2. The method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud according to claim 1 is characterized in that: It also includes step 6: doubt elimination, labeling and visual output, which conducts human-computer interaction verification and elimination on the doubts detected automatically, and determines whether the doubts are valid by checking relevant evidence or records. If it is found that the doubts are caused by design changes, data entry errors or other normal business activities, these doubts will be eliminated.

3. The method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud according to claim 1 is characterized in that: The step 1 comprises: Step 1-1: Obtain relevant content of the project through business systems, regulatory platforms and Internet channels, conduct data governance, record key time nodes of the project, and convert the relevant content and key time node information into structured data for storage as project feature input; Step 1-2: Use the 3D model built using BIM software during the project design phase to add component material information based on the design instructions and construction specifications; Step 1-3: Collect laser point cloud data from the early stage of the project as the training data source for the semantic segmentation model, record and save the results of multiple laser point cloud phases, and form a complete data chain.

4. The method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud according to claim 1 is characterized in that: The step 2 comprises: Step 2-1: Clarify the projection coordinate system used in the project design and the geographic coordinate system to be converted; Step 2-2: Set the plane coordinate information of the model data through BIM software to determine its latitude, longitude and altitude; Step 2-3: Obtain the angle between the project north and true north through drawing measurement, which is used as the angle between the BIM model and true north; Step 2-4: Use conversion tools or conversion algorithms to convert the BIM model from planimetric coordinates to geospatial coordinates.

5. The method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud according to claim 1 is characterized in that: The step 3 comprises: Step 3-1: Filter the key component categories involved in the audit and detection based on semantic information, extract the corresponding BIM model components and convert them into triangular surface models, perform topological structure correction and data compression after conversion, reduce the number of triangles, and reduce the complexity of the model; Step 3-2: Determine the sampling point density, capture the curvature changes of the geometric surface, and realize the three-dimensional point cloud expression of the BIM model components through the closed contour point cloud filling method; Step 3-3: Assign sampling point attributes, associate the sampling points with project management information, and export them to a standard point cloud file format.

6. The method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud according to claim 1 is characterized in that: In step 4, pre-training includes data set preparation and model training, and the specific steps are as follows: Step 4-1: Manually annotate the point cloud data collected in the early stage, generate point cloud samples covering important categories of components and their corresponding semantic labels, and integrate them into the existing 3D public dataset to form a high-quality training dataset covering a variety of scenarios; Step 4-2: Consider the accuracy of local feature extraction and the efficiency of global semantic segmentation, and select an appropriate training model framework; Step 4-3: Preprocess the point cloud data and divide the data set into training set, validation set and test set; Step 4-4: Define the loss function and optimization strategy, optimize the model step by step, adjust the hyperparameters using the validation set, improve the generalization ability and robustness of the BIM model, and evaluate the BIM model using the test set.

7. The method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud according to claim 1 is characterized in that: In step 5, the contents of the inspection include: inspecting position and / or size deviation, inspecting material deviation, and inspecting unfinished parts.

8. The method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud according to claim 7 is characterized in that: The position and / or size deviation detection method is to determine whether there is a position deviation by calculating the symmetrical distance between two clusters of point clouds of the same component in the adjacent space. When the distance exceeds a set threshold, it can be determined that the component position has deviated. The specific method is as follows: Step a: Assume two clusters of point clouds, where point cloud A contains the point set , point cloud B contains the point set , for each point in point cloud A , find the nearest point in point cloud B , and calculate its distance. Similarly, for each point in point cloud B , find the nearest point in point cloud A ; Step b: The average closest point distance from point cloud A to point cloud B: , where m is a constant; Step c: The average closest point distance from point cloud B to point cloud A: , where n is a constant; Step d: Symmetric distance calculation: 。 9. The method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud according to claim 7 is characterized in that: The material deviation detection includes extracting the color features of each component in the measured point cloud, and searching for the corresponding BIM model component point cloud based on its spatial position to obtain the corresponding material semantic information, and comparing it with the point cloud data of the same material label in the training data set. The specific steps are as follows: Step e: discretize the color value into multiple color intervals, calculate the number or frequency of point clouds in each color interval, and obtain a color histogram as a description of the color feature; Step f: Use the distance metric formula Compare the color histogram differences of the two groups of point clouds to quantify the material deviation, where: and Respectively represent the histogram values ​​of point cloud A and point cloud B in the color range.

10. The method for extracting doubtful points in construction project engineering management audit based on three-dimensional point cloud according to claim 7, characterized in that: The unfinished parts of the inspection include: Step g: Filter the components that should have been built at the current time in the BIM model point cloud by combining the construction period time semantics, and search for the measured point cloud that matches it based on the spatial position; Step h: quantify the coverage of the point clouds by calculating the intersection-and-union ratio between the two, voxelize the two point clouds respectively, assign each point to the corresponding voxel, and calculate the intersection and union of the two sets of voxel sets. The overlap is expressed as: , when the calculation result is lower than a certain threshold, it can be determined that the component is not completed.

Citation Information

Patent Citations

  • Automatic reconstruction method for pipeline facility BIM model based on point cloud data

    CN108898662A

  • Existing structure BIM model automatic generation method based on three-dimensional point cloud

    CN113868733A

  • Visual early warning method for water conveyance tunnel diseases

    CN114494385A

  • Subway station construction quality evaluation method based on laser scanning technology

    CN115018249A

  • Pipeline BIM model reconstruction method and system based on depth image and point cloud

    CN115294294A

Cited By

  • Building structure component comparison method and system oriented to project supervision

    CN121121643A

  • A method and system for comparing building structural components for engineering supervision

    CN121121643B

  • High-precision building structure construction deviation layered detection method based on forward modeling

    CN121366118A

  • House building main body frame measuring method, device and equipment

    CN121482134A