A construction error detection method and system based on three-dimensional laser scanning
Through three-dimensional laser scanning technology, building point cloud data is collected and processed, combined with machine learning to identify construction errors and generate correction solutions, the problems of low data processing efficiency and inaccurate defect identification in traditional methods are solved, and efficient and accurate construction error correction is achieved. , improves construction quality and efficiency.
Patent Information
- Application Number
- CN202411228219.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-09-03
AI Technical Summary
The existing building construction error detection methods have problems such as low data processing efficiency, inaccurate defect identification, incomplete defect evaluation, and insane repair plan formulation. In particular, traditional methods rely on manual inspections, resulting in time-consuming and limited reliability.
Using a three-dimensional laser scanning method, the point cloud data of building structures is collected by deploying high-performance sensors, preprocessing and three-dimensional modeling, and the model is constructed using point cloud clustering and surface fitting algorithms to identify potential errors compared with the construction standard model. To build a defect identification model in combination with CNN, generate defect reports, and automatically plan correction plans through the construction management platform to ensure the correction effect.
It realizes efficient data collection and processing, accurately identify construction errors, improves the accuracy of defect identification and correction efficiency, and improves the quality and economicality of construction.
Smart Images

Figure CN118735922B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of building construction error detection, and in particular to a building construction error detection method and system based on three-dimensional laser scanning. Background Art
[0002] With the rapid development of information technology, 3D laser scanning technology, as an advanced non-contact measurement method, has been widely used in civil engineering, architectural design and other fields. This technology can quickly obtain the 3D coordinate information of the surface of an object, generate high-precision point cloud data, and provide detailed spatial information for construction. In recent years, with the advancement of sensor technology and the enhancement of computing power, 3D laser scanning technology has not only significantly improved the scanning speed, but also made great progress in data processing algorithms. These technological advances have made 3D laser scanning an important tool that can be used in many scenarios such as construction error detection and deformation monitoring.
[0003] However, although 3D laser scanning technology has achieved automation and intelligence to a certain extent, there are still some challenges in practical application. For example, how to efficiently process massive point cloud data, how to accurately build 3D models and compare them with construction standard models, and how to effectively identify construction defects are still urgent issues to be solved. In addition, traditional construction error detection methods often rely on manual inspection, which is not only time-consuming and labor-intensive, but also prone to human omissions, resulting in limited reliability of detection results. Summary of the invention
[0004] In view of the above existing problems, the present invention is proposed.
[0005] Therefore, the present invention provides a construction error detection method and system based on three-dimensional laser scanning to solve the problems of low data processing efficiency, inaccurate defect identification, incomplete defect assessment and unscientific repair plan formulation.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:
[0007] In a first aspect, the present invention provides a method for detecting construction errors based on three-dimensional laser scanning, comprising: deploying three-dimensional laser scanning equipment to collect point cloud data of a building structure, and preprocessing the point cloud data; constructing a first three-dimensional model of the building structure based on the preprocessed point cloud data, and comparing and analyzing the first three-dimensional model with a construction standard model to identify potential construction error points; constructing a three-dimensional defect recognition model based on the potential construction error points, identifying construction defects and generating a construction defect report; transmitting the construction defect report to a construction management platform, and the construction management platform plans and executes a correction plan based on the construction defect report; after the construction defect correction is completed, scanning the point cloud data of the corrected building structure, constructing a second three-dimensional model of the corrected building structure, and comparing the second three-dimensional model with the construction standard model for a second time to verify the correction effect.
[0008] As a preferred solution of the construction error detection method based on three-dimensional laser scanning of the present invention, the three-dimensional laser scanning equipment is deployed to collect point cloud data of the building structure, and the point cloud data is preprocessed including:
[0009] 3D laser scanners deploying high-performance sensors;
[0010] Use a 3D laser scanner to scan the building structure in all directions to obtain point cloud data including X, Y, and Z coordinates and reflectivity;
[0011] Transmit point cloud data to a central monitoring computer system;
[0012] Use outlier detection algorithms to identify and remove points that deviate from the set standard distribution;
[0013] Perform median filtering on the point cloud data, retain the points within the preset neighborhood of the median, and remove extreme values;
[0014] PCL is used to convert the preprocessed point cloud data into a unified format.
[0015] As a preferred solution of the construction error detection method based on three-dimensional laser scanning of the present invention, the first three-dimensional model of the building structure is constructed based on the preprocessed point cloud data, and the first three-dimensional model is compared and analyzed with the construction standard model to identify potential construction error points, including:
[0016] Import the preprocessed point cloud data into CloudCompare, use the point cloud clustering algorithm to extract the key features of the building structure, and use the surface fitting algorithm to build the first three-dimensional model of the building structure;
[0017] Retrieve the construction standard model based on design drawings and construction specifications from the database;
[0018] The least squares algorithm is used to calculate the deviation between the coordinates of each point in the point cloud data of the first 3D model and the coordinates of the corresponding point in the construction standard model. The expression is: ; Where N represents the total number of points in the point cloud data, Represents the average standard deviation of all point deviations in the point cloud data, represents the Euclidean distance between the i-th point and the corresponding point in the construction standard model, and Q represents the deviation quantification index;
[0019] Defining deviation quantification limit thresholds based on construction specifications , according to the deviation quantification index Deviation quantification limit threshold Compare the results of the above analysis to identify potential error points with large deviations;
[0020] when When , it is considered that there are potential construction error points in the current area;
[0021] when When , it is considered that the current area meets the construction standards and there is no potential error point;
[0022] Deviations exceeding the deviation quantification limit threshold will be identified The point cloud data is marked as potential construction error points.
[0023] As a preferred solution of the construction error detection method based on three-dimensional laser scanning of the present invention, the three-dimensional defect recognition model is constructed based on potential construction error points, including:
[0024] Based on the historical building defect dataset, an initial 3D defect recognition model was constructed using CNN as the infrastructure.
[0025] Define a loss function to measure the gap between the model prediction and the actual label, and use the cross entropy loss function to measure the performance of the classification task;
[0026] Update model parameters according to the gradient of the loss function;
[0027] The point cloud data of potential error points is input into the updated 3D defect recognition model to identify the probability of various defect types. The expression is: ;
[0028] in, Indicates The probability of a defect type, Indicates The classification score corresponding to the defect, K represents the total number of defect categories;
[0029] Based on defect type probability , construct a severity evaluation function to evaluate the severity of the defect, the expression is:
[0030] ;in, It is a comprehensive feature vector that integrates the distance, texture, shape and context features of each point in the point cloud data. Indicates the severity of the defect.
[0031] As a preferred solution of the construction error detection method based on three-dimensional laser scanning of the present invention, the identification of construction defects and generation of construction defect reports include:
[0032] Defining standard thresholds for defect repair based on safety requirements for building construction , the severity of the defect Defect repair standard threshold Make comparisons and identify defects that need to be corrected;
[0033] when When , it is considered that the severity of the defect point exceeds the normal range and needs to be corrected;
[0034] when When the defect is within the normal range, it is considered that the severity of the defect is within the normal range and no correction is required;
[0035] The probability of defect type , the location of defect points and the information of identified defect points that need to be corrected are summarized into a construction defect report.
[0036] As a preferred solution of the construction error detection method based on three-dimensional laser scanning of the present invention, the construction defect report is transmitted to the construction management platform, and the construction management platform plans and executes a correction plan according to the construction defect report, including:
[0037] Transmit construction defect reports to the construction management platform, using encryption protocols to ensure the security of data transmission;
[0038] The construction management platform uses a decision tree to generate preliminary correction plans based on the content in the construction defect report;
[0039] The generated preliminary correction scheme is optimized by genetic algorithm;
[0040] Based on the correction cost, defect severity and estimated correction time, the following defect correction plan evaluation function is used to evaluate the optimized correction plan. The expression is: ;
[0041] in, Indicates the suitability of the correction plan. represents the correction cost, Indicates the estimated time of correction. is the importance coefficient of cost, is the importance coefficient of severity, is the importance coefficient of the correction time;
[0042] According to suitability An optimal defect correction scheme is selected, and a correction instruction for the optimal defect correction scheme is generated, and the correction instruction is sent to a mobile terminal of a construction worker, wherein the correction instruction is used to instruct the construction worker to execute the optimal defect correction scheme according to the correction instruction.
[0043] As a preferred solution of the construction error detection method based on three-dimensional laser scanning of the present invention, after the construction defect correction is completed, the point cloud data of the corrected building structure is scanned, a second three-dimensional model of the corrected building structure is constructed, and the second three-dimensional model is compared with the construction standard model for a second time to verify the correction effect, which includes:
[0044] After the construction defect correction is completed, the corrected building structure is scanned again to obtain new point cloud data including X, Y, and Z coordinates and new reflectivity, and a second three-dimensional model of the corrected building structure is reconstructed;
[0045] The second 3D model is compared with the construction standard model for a second time to verify whether the correction effect meets the expectations. If there are still parts that do not meet the standards, the system will start the defect identification process again until all construction errors are corrected to within the allowable range.
[0046] In the second aspect, the present invention provides a construction error detection system based on three-dimensional laser scanning, including: a data acquisition and preprocessing module, a three-dimensional modeling and comparison module, a defect identification and reporting module, a construction management and correction module, and a verification and feedback module, wherein the data acquisition and preprocessing module is used to deploy three-dimensional laser scanning equipment to collect point cloud data of the building structure, and preprocess the collected point cloud data; the three-dimensional modeling and comparison module is used to construct a first three-dimensional model of the building structure based on the preprocessed point cloud data, compare and analyze the first three-dimensional model with the construction standard model, and identify potential construction error points; the defect identification and reporting module is used to construct a three-dimensional defect identification model based on the potential construction error points, identify construction defects and generate a construction defect report; the construction management and correction module is used to transmit the construction defect report to the construction management platform, and plan the correction plan according to the construction defect report; the verification and feedback module is used to compare the second three-dimensional model of the corrected building structure with the construction standard model for a second time after the construction defect correction is completed to verify the correction effect.
[0047] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, any step of the construction error detection method based on three-dimensional laser scanning as described in the first aspect of the present invention is implemented.
[0048] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any step of the construction error detection method based on three-dimensional laser scanning as described in the first aspect of the present invention.
[0049] The beneficial effects of the present invention are as follows: by deploying a 3D laser scanner with a high-performance sensor to collect point cloud data of the building structure and preprocessing it, efficient data collection and high-quality processing are achieved; then, a 3D model of the building structure is constructed based on the preprocessed point cloud data and compared and analyzed with the construction standard model, accurate identification of potential construction error points is achieved; then, a 3D defect recognition model is constructed to identify the defect type and severity of the error points, and a defect report is generated, thereby ensuring the accuracy and completeness of defect identification; finally, the construction management platform automatically plans and executes the correction plan, which realizes the effective correction of construction errors and improves the quality and efficiency of construction. The present invention effectively solves the error problem in construction and improves the accuracy and economy of construction. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use 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 ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0051] Figure 1 This is a flow chart of the construction error detection method based on three-dimensional laser scanning in Example 1.
[0052] Figure 2 This is a flow chart of the construction error detection system based on three-dimensional laser scanning in Example 1. DETAILED DESCRIPTION
[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.
[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0055] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.
[0056] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0057] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a construction error detection method based on three-dimensional laser scanning, comprising:
[0058] S1: Deploy 3D laser scanning equipment to collect point cloud data of building structures and pre-process the point cloud data.
[0059] Furthermore, at the construction site, a 3D laser scanner with high-performance sensors is deployed. This device can capture point cloud data of building structures with centimeter-level accuracy, ensuring high data density and accuracy.
[0060] Use a 3D laser scanner to scan the entire building structure in all directions to obtain point cloud data containing X, Y, and Z coordinates as well as additional attributes such as reflectivity;
[0061] Transmit point cloud data to the central monitoring computer system in real time via high-speed wireless network to ensure the security and efficiency of data transmission;
[0062] Use outlier detection algorithms to identify and delete points that deviate from the set standard distribution to reduce the impact of noise;
[0063] The specific process is as follows: first, statistical or neighborhood analysis methods are applied to the original point cloud data obtained by the 3D laser scanner to identify points that deviate from the set standard distribution; then, by calculating the distance from each point to its nearest neighbor or indicators such as local density, it is determined which points may be outliers; finally, according to the set standards, these abnormal points are removed to reduce the impact of noise on subsequent processing steps and improve the quality and availability of point cloud data.
[0064] Perform median filtering on the point cloud data to retain points within the preset range of the median, remove extreme values, and reduce random noise;
[0065] The specific process is: first, select each point in the point cloud as the center point and define a neighborhood range around it; then, select a certain number of nearest neighboring points in the neighborhood; then, calculate the median of these neighborhood points as the new value of the center point; finally, repeat this process until all points have been processed, thereby retaining points within the preset range of the median, effectively removing extreme values, reducing the impact of random noise on point cloud data, and improving the smoothness and accuracy of the data.
[0066] PCL is used to convert the pre-processed point cloud data into a unified format for subsequent software processing.
[0067] The specific process is as follows: first, load the pre-processed point cloud data into the PCL environment; then, use the functions or tools provided by PCL to convert the point cloud data into a standard format supported by PCL, such as pcl; then, make necessary format adjustments to the converted point cloud data to ensure data consistency and compatibility; finally, save the converted point cloud data for further processing and analysis by subsequent software or algorithms. Through this process, it can be ensured that the point cloud data can interact smoothly between different software and provide a consistent data format for subsequent 3D modeling and analysis.
[0068] S2: Based on the preprocessed point cloud data, a first three-dimensional model of the building structure is constructed, and the first three-dimensional model is compared and analyzed with the construction standard model to identify potential construction error points.
[0069] Furthermore, the pre-processed point cloud data is imported into CloudCompare, and the key features of the building structure are extracted using the point cloud clustering algorithm, wherein the key features may include planes, curved surfaces, edges and texture features, etc., and the first three-dimensional model of the building structure is constructed using the surface fitting algorithm;
[0070] It should be noted that the specific process of extracting the key features of the building structure using the point cloud clustering algorithm is as follows: first, the preprocessed point cloud data is imported into professional software such as CloudCompare; then, the point cloud data is segmented using the point cloud clustering algorithm, and similar point cloud data are clustered together to form different clusters; then, feature extraction is performed on the point cloud data within each cluster, such as basic geometric features such as planes, columns or surfaces; finally, a three-dimensional model of the building structure is constructed through these key features, laying the foundation for subsequent comparative analysis with the construction standard model.
[0071] The specific process of using the surface fitting algorithm to construct a three-dimensional model of the building structure is as follows: first, the pre-processed point cloud data is imported into the professional software CloudCompare; then, the point cloud data is segmented using the point cloud clustering algorithm, and similar point cloud data are clustered together to form different clusters; then, the point cloud data in each cluster is subjected to surface fitting processing, and the best fitting surface is found through a fitting algorithm (such as the least squares method) to describe the distribution of the point cloud data in each cluster; finally, all the fitted surfaces are combined to construct a complete three-dimensional model of the building structure.
[0072] Retrieve the construction standard model based on design drawings and construction specifications from the database, which defines the ideal geometric parameters and allowable tolerances;
[0073] It should also be noted that design drawings and construction specifications are usually prepared by architects and engineers based on project needs, customer requirements, and relevant laws and regulations. Design drawings and construction specifications are used as reference benchmarks and are usually stored in the project database for easy reference when building 3D models and comparative analysis. These documents are ready at the beginning of the project and play a guiding and supervisory role throughout the construction process.
[0074] Using the least squares method and other algorithms, calculate the deviation between the coordinates of each point in the point cloud data of the first 3D model and the coordinates of the corresponding point in the construction standard model. The expression is: ; Where N represents the total number of points in the point cloud data, Represents the average standard deviation of all point deviations in the point cloud data, represents the Euclidean distance between the i-th point and the corresponding point in the construction standard model, and Q represents the deviation quantification index;
[0075] It should be noted that the Euclidean distance between the measured point and the corresponding point in the standard model is , the expression is: ;in, Represents the X coordinate value of the i-th point in the point cloud data, Represents the Y coordinate value of the i-th point in the point cloud data, Represents the Z coordinate value of the i-th point in the point cloud data, shows the X coordinate of the corresponding measurement point in the three-dimensional standard model, Indicates the Y coordinate of the corresponding measurement point in the three-dimensional standard model, Indicates the Z coordinate of the corresponding measurement point in the three-dimensional standard model;
[0076] Define a deviation quantification limit threshold based on construction specifications , according to the deviation quantification index Deviation quantification limit threshold Compare the results of the above analysis to identify potential error points with large deviations;
[0077] when When , it is considered that there are potential construction error points in the current area;
[0078] when When , it is considered that the current area meets the construction standards and there is no potential error point;
[0079] Deviations exceeding the deviation quantification limit threshold will be identified The point cloud data is marked as potential construction error points (marked in the form of three-dimensional coordinates).
[0080] S3: Build a 3D defect recognition model based on potential construction error points, identify construction defects and generate a construction defect report.
[0081] Furthermore, based on the historical building defect dataset, including various defect types such as cracks, unevenness, and misalignment, CNN is used as the basic architecture because CNN is good at processing spatial features in images and point cloud data to build an initial three-dimensional defect recognition model;
[0082] Define the loss function to measure the gap between the model prediction and the actual label, and use the cross entropy loss function to measure the performance of the classification task. The expression is: ;in, represents the actual label of the jth sample (such as crack size, unevenness, etc.), It indicates the probability that the model predicts that the jth sample belongs to a specific category, M indicates the number of samples in a batch, and L indicates the loss function value;
[0083] Update the model parameters according to the gradient of the loss function. The specific steps are: first, define the loss function to measure the gap between the model prediction and the actual label; then, calculate the gradient of the loss function with respect to the model parameters, that is, the rate of change of the loss function relative to the model parameters; then, use optimization algorithms such as gradient descent to adjust the model parameters according to the direction of the gradient to reduce the value of the loss function; finally, repeat this process until the loss function converges to a smaller value, thereby obtaining the optimized model parameters.
[0084] The point cloud data of potential error points is input into the updated and optimized 3D defect recognition model to identify the probability of various defect types, such as cracks, unevenness, dislocation, etc., taking into account the size and location of the defects. The expression is: ;in, Indicates The probability of a defect type, Indicates The classification score corresponding to the defect, K represents the total number of defect categories;
[0085] It should be noted that The expression is: ;in, represents the weight vector of the h-th defect type, Represents the feature vector input to the last layer, which is usually the result of processing by convolution layer, activation function, pooling layer, etc. represents the bias vector;
[0086] Based on defect type probability , construct a severity evaluation function to evaluate the severity of the defect, the expression is: ;in, It is a comprehensive feature vector that integrates the distance, texture, shape and context features of each point in the point cloud data. Indicates the severity of the defect;
[0087] It should be noted that the feature fusion function Combining the distance, texture features, and shape features of each point in the point cloud data, the expression is: ; Where D is the average value of the Euclidean distance between points in the point cloud, T is the texture feature vector, S is the shape feature vector, is the importance coefficient of the specific feature, is the texture importance coefficient, γ is the shape importance coefficient;
[0088] Define a standard threshold for defect repair based on safety requirements for building construction , the severity of the defect Defect repair standard threshold Make comparisons and identify defects that need to be corrected;
[0089] when When , it is considered that the severity of the defect point exceeds the normal range and needs to be corrected;
[0090] when When the defect is within the normal range, it is considered that the severity of the defect is within the normal range and no correction is required;
[0091] For example, suppose the severity of a potential error point has been calculated = 0.75, and the severity threshold has been set =0.5.
[0092] Comparison process: 0.75>0.5.
[0093] Conclusion: Because the severity of the potential error point exceeds the set defect repair standard threshold, the potential error point needs to be corrected.
[0094] The probability of defect type , defect point location and identified defect points that need to be corrected are summarized into a defect report. The defect report includes: defect type: such as cracks, unevenness, misalignment, etc.;
[0095] Defect location: including X, Y, and Z coordinate values;
[0096] Defect severity: evaluated by severity function Calculated.
[0097] S4: The construction defect report is transmitted to the construction management platform, and the construction management platform plans and executes the correction plan based on the construction defect report.
[0098] Furthermore, the construction defect reports are transmitted to the construction management platform in real time via a secure wireless network, using encryption protocols to ensure the security of data transmission;
[0099] It should be noted that the process of using encryption protocols to ensure the security of data transmission is as follows: first, in the process of transmitting construction defect reports from on-site equipment to the construction management platform, TLS / SSL is used to encrypt the data; second, before data transmission, both parties confirm the encryption key to ensure the security of the data during transmission; then, the original data is converted into ciphertext through an encryption algorithm. Even if the data is intercepted during transmission, it cannot be interpreted by a third party; finally, when the data arrives at the construction management platform, the corresponding decryption key is used to restore the ciphertext to the original data to ensure the integrity and confidentiality of the data.
[0100] The construction management platform uses a decision tree to generate preliminary correction plans based on the content in the construction defect report;
[0101] The specific process of generating a preliminary correction plan from a decision tree is as follows: first, the platform extracts key information from the report, such as defect type, location, severity, etc.; then, it makes judgments in the preset decision tree model based on this information, and gradually traverses the nodes downward. Each node represents a decision point, such as whether it is a structural problem, whether it affects progress, etc.; next, based on the answer to each decision point, the decision tree will lead to different branch paths and eventually reach a leaf node, which contains corrective measures recommended for specific situations; finally, the platform integrates all suggestions to generate a complete correction plan and feeds it back to relevant personnel for execution.
[0102] The generated preliminary correction scheme is optimized by genetic algorithm to reduce the cost, shorten the correction time or improve the correction efficiency;
[0103] The specific process of optimizing the preliminary correction scheme through genetic algorithm is as follows: first, convert the preliminary correction scheme into a coding form suitable for the operation of genetic algorithm; then, set the initial population, each individual represents a different version of the correction scheme; then, iteratively evolve the population through genetic operations such as selection, crossover and mutation, among which the selection operation retains the scheme with better performance, the crossover operation combines the advantages of different schemes to generate new schemes, and the mutation operation randomly adjusts certain parts of the scheme to introduce new solutions; after multiple rounds of iterations, the algorithm will gradually converge to a better correction scheme, thereby improving the efficiency and effectiveness of the correction work.
[0104] Based on the correction cost, defect severity and estimated correction time, a defect correction plan evaluation function is designed to evaluate the optimized correction plan. The expression is: ;in, Indicates the suitability of the correction plan. represents the correction cost, Indicates the estimated time of correction. is the importance coefficient of cost, is the importance coefficient of severity, is the importance coefficient of the correction time;
[0105] when When it is close to 0, it means that the cost of the correction plan is high, the severity is low, or the estimated correction time is long, and the overall suitability is poor;
[0106] when When it is close to 1, it means that the cost of the correction plan is low, the severity is high, or the estimated correction time is short, and the overall suitability is good.
[0107] According to suitability After selecting the best defect correction plan, the construction management platform generates detailed correction instructions, including specific correction methods, required tools and materials, etc., and sends the correction instructions to the construction personnel's mobile terminals. After receiving the correction instructions, the construction personnel begin to implement the best defect correction plan.
[0108] S5: After the construction defect correction is completed, the point cloud data of the corrected building structure is scanned, a second three-dimensional model of the corrected building structure is constructed, and the second three-dimensional model is compared with the construction standard model for a second time to verify the correction effect.
[0109] Furthermore, after the construction defect correction is completed, the corrected building structure is scanned again to obtain new point cloud data including X, Y, and Z coordinates and additional attributes such as new reflectivity, and a second three-dimensional model of the corrected building structure is reconstructed;
[0110] It should be noted that the acquired new point cloud data and new reflectivity, etc., must be processed through the preprocessing process in S1;
[0111] The second three-dimensional model of the corrected building structure is compared with the construction standard model for a second time to verify whether the correction effect meets the expected standards. If there are still parts that do not meet the standards, the system will start the defect identification process again until all construction errors are corrected to within the standard range.
[0112] It should be noted that the expected standard refers to the ideal state specified in the design drawings, including the geometric size, shape, position, etc. of the building structure. The standard range is the allowable deviation value defined in the construction specifications, that is, the maximum deviation limit allowed during the construction process, to ensure that the building can meet the safety and functionality of the design requirements in actual construction.
[0113] It should also be noted that the defect identification process refers to steps S2 to S3 in this embodiment;
[0114] The revised 3D model and verification results are fed back to the construction management platform to update the project progress and quality control database, while providing reference and guidance for subsequent construction.
[0115] The present embodiment also provides a construction error detection system based on 3D laser scanning, including: a data acquisition and preprocessing module, a 3D modeling and comparison module, a defect identification and reporting module, a construction management and correction module, and a verification and feedback module, wherein the data acquisition and preprocessing module is used to deploy 3D laser scanning equipment to collect point cloud data of the building structure and preprocess the collected point cloud data; the 3D modeling and comparison module is used to construct a first 3D model of the building structure based on the preprocessed point cloud data, compare and analyze the first 3D model with the construction standard model, and identify potential construction error points; the defect identification and reporting module is used to construct a 3D defect identification model based on the potential construction error points, identify construction defects and generate a construction defect report; the construction management and correction module is used to transmit the construction defect report to the construction management platform, and plan the correction plan according to the construction defect report; the verification and feedback module is used to compare the second 3D model of the corrected building structure with the construction standard model for a second time after the construction defect correction is completed to verify the correction effect.
[0116] This embodiment also provides a computer device, which is suitable for the construction error detection method based on three-dimensional laser scanning, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the construction error detection method based on three-dimensional laser scanning proposed in the above embodiment.
[0117] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.
[0118] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for detecting construction errors based on three-dimensional laser scanning as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.
[0119] In summary, the present invention achieves high-precision data capture and high-quality data processing by: deploying high-performance three-dimensional laser scanning equipment to collect point cloud data of building structures and preprocessing, thereby constructing an accurate three-dimensional model and comparing and analyzing it with the construction standard model to identify potential construction error points. Advanced machine learning technology is used to build a three-dimensional defect recognition model to automatically identify construction defects and assess their severity, and generate a construction defect report. The construction management platform automatically plans and optimizes correction plans based on these reports to ensure that construction defects are corrected efficiently and accurately. The correction effect is verified through secondary comparison to ensure that all construction errors are corrected within the standard range, thereby improving construction quality and efficiency.
[0120] Example 2, referring to Table 1, is the second example of the present invention. To further verify the advancement of the present invention, experimental simulation data of a construction error detection method based on three-dimensional laser scanning is provided.
[0121] Deploy a high-performance 3D laser scanner at the construction site. The device has centimeter-level accuracy and captures high-density and accurate point cloud data of the building structure. Before scanning, ensure that all sensors are calibrated to ensure data accuracy.
[0122] First, high-precision 3D laser scanning equipment was deployed to collect point cloud data of the building structure, and necessary preprocessing was performed to ensure the accuracy and usability of the data.
[0123] Secondly, the data was deeply analyzed using advanced point cloud processing algorithms, including outlier detection and median filtering, to identify deviations from the construction standard model.
[0124] Next, a deep learning-based 3D defect recognition model was built to evaluate the type and severity of defects and generate correction plans.
[0125] Finally, the correction plan was optimized and implemented through the decision support system of the construction management platform, and the correction effect was verified by re-scanning to ensure that the construction quality met the design and specification requirements.
[0126] The details are shown in Table 1:
[0127] Table 1 Building defect analysis table
[0128]
[0129] By analyzing the data in the above table, we can see the effectiveness and superiority of the method of the present invention. For example, it can be seen from the table that the deviation quantification index Q of building B and building D is 0.006 and 0.008 respectively, both exceeding the deviation quantification limit threshold. =0.005, which indicates that there are potential error points in these two building areas. The deviation quantification index Q of other buildings (building A, building C, building E) is lower than , indicating that there are no obvious construction errors in these areas. This shows that the method of the present invention can effectively identify the building areas that need to be corrected, which helps to take timely measures to correct them and avoid potential safety hazards.
[0130] The construction error detection method based on 3D laser scanning of the present invention can not only improve the quality control level of construction, ensure the safety and durability of the building structure, but also reduce construction costs and improve construction efficiency. Especially in the face of complex and large-scale construction projects, the method of the present invention can play a greater role, ensuring that every link in the construction process can be effectively supervised and improved.
[0131] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.
Claims
1. A construction error detection method based on three-dimensional laser scanning, characterized in that: include: Deploy 3D laser scanning equipment to collect point cloud data of building structures and pre-process the point cloud data; Based on the pre-processed point cloud data, a first three-dimensional model of the building structure is constructed, and the first three-dimensional model is compared and analyzed with the construction standard model to identify potential construction error points; Build a 3D defect recognition model based on potential construction error points to identify construction defects and generate a construction defect report; Transmit the construction defect report to the construction management platform, which plans and executes the correction plan based on the construction defect report; After the construction defect correction is completed, the point cloud data of the corrected building structure is scanned, and a second 3D model of the corrected building structure is constructed. The second 3D model is compared with the construction standard model to verify the correction effect; Deploying a 3D laser scanning device to collect point cloud data of the building structure and preprocessing the point cloud data includes: 3D laser scanners deploying high-performance sensors; Use a 3D laser scanner to scan the building structure in all directions to obtain point cloud data including X, Y, and Z coordinates and reflectivity; Transmit point cloud data to a central monitoring computer system; Use outlier detection algorithms to identify and remove points that deviate from the set standard distribution; Perform median filtering on the point cloud data, retain the points within the preset neighborhood of the median, and remove extreme values; Use PCL to convert the pre-processed point cloud data into a unified format; The method of constructing a first three-dimensional model of the building structure based on the pre-processed point cloud data, comparing and analyzing the first three-dimensional model with the construction standard model, and identifying potential construction error points includes: Import the preprocessed point cloud data into CloudCompare, use the point cloud clustering algorithm to extract the key features of the building structure, and use the surface fitting algorithm to build the first three-dimensional model of the building structure; Retrieve the construction standard model based on design drawings and construction specifications from the database; The least squares method is used to calculate the deviation between the coordinates of each point in the point cloud data of the first 3D model and the coordinates of the corresponding point in the construction standard model. The expression is: , Where N represents the total number of points in the point cloud data. Represents the average standard deviation of all point deviations in the point cloud data, represents the Euclidean distance between the i-th point and the corresponding point in the construction standard model, and Q represents the deviation quantification index; Defining deviation quantification limit thresholds based on construction specifications , according to the deviation quantification index Deviation quantification limit threshold Compare the results of the above analysis to identify potential error points with large deviations; when > When , it is considered that there are potential construction error points in the current area; when ≤ When , it is considered that the current area meets the construction standards and there is no potential error point; Deviations exceeding the deviation quantification limit threshold will be identified Point cloud data is marked as potential construction error points; The construction of a three-dimensional defect recognition model based on potential construction error points includes: Based on the historical building defect dataset, an initial 3D defect recognition model was constructed using CNN as the infrastructure. Define a loss function to measure the gap between the model prediction and the actual label, and use the cross entropy loss function to measure the performance of the classification task; Update model parameters according to the gradient of the loss function; The point cloud data of potential error points is input into the updated 3D defect recognition model to identify the probability of various defect types. The expression is: , in, Indicates The probability of a defect type, Indicates The classification score corresponding to the defect, K represents the total number of defect categories; Based on defect type probability , construct a severity evaluation function to evaluate the severity of the defect, the expression is: , in, It is a comprehensive feature vector that integrates the distance, texture, shape and context features of each point in the point cloud data. Indicates the severity of the defect; The identifying of construction defects and generating a construction defect report comprises: Defining standard thresholds for defect repair based on safety requirements for building construction , the severity of the defect Defect repair standard threshold Make comparisons and identify defects that need to be corrected; when > When , it is considered that the severity of the defect point exceeds the normal range and needs to be corrected; when ≤ When the defect is within the normal range, it is considered that the severity of the defect is within the normal range and no correction is required; The probability of defect type , the location of defect points and the information of defect points that need to be corrected are summarized into a construction defect report; The construction defect report is transmitted to the construction management platform, and the construction management platform plans and executes a correction plan based on the construction defect report, including: Transmit construction defect reports to the construction management platform, using encryption protocols to ensure the security of data transmission; The construction management platform uses a decision tree to generate preliminary correction plans based on the content in the construction defect report; The generated preliminary correction scheme is optimized by genetic algorithm; Based on the correction cost, defect severity and estimated correction time, the following defect correction plan evaluation function is used to evaluate the optimized correction plan. The expression is: , in, Indicates the suitability of the correction plan. represents the correction cost, Indicates the estimated time of correction. is the importance coefficient of cost, is the importance coefficient of severity, is the importance coefficient of the correction time; According to suitability An optimal defect correction scheme is selected, and a correction instruction for the optimal defect correction scheme is generated, and the correction instruction is sent to a mobile terminal of a construction worker, wherein the correction instruction is used to instruct the construction worker to execute the optimal defect correction scheme according to the correction instruction.
2. The construction error detection method based on three-dimensional laser scanning according to claim 1, characterized in that: After the construction defect correction is completed, the point cloud data of the corrected building structure is scanned, a second three-dimensional model of the corrected building structure is constructed, and the second three-dimensional model is compared with the construction standard model for a second time to verify the correction effect, including: After the construction defect correction is completed, the corrected building structure is scanned again to obtain new point cloud data including X, Y, and Z coordinates and new reflectivity, and a second three-dimensional model of the corrected building structure is reconstructed; The second 3D model is compared with the construction standard model for a second time to verify whether the correction effect meets the expectations. If there are still parts that do not meet the standards, the system will start the defect identification process again until all construction errors are corrected to within the allowable range.
3. A construction error detection system based on 3D laser scanning, based on the construction error detection method based on 3D laser scanning according to claim 1 or 2, characterized in that: include: Data acquisition and preprocessing module, 3D modeling and comparison module, defect identification and reporting module, construction management and correction module and verification and feedback module, among which, The data acquisition and preprocessing module is used to deploy 3D laser scanning equipment to collect point cloud data of building structures and preprocess the collected point cloud data; The 3D modeling and comparison module is used to construct a first 3D model of the building structure based on the pre-processed point cloud data, compare and analyze the first 3D model with the construction standard model, and identify potential construction error points; The defect identification and reporting module is used to build a three-dimensional defect identification model based on potential construction error points, identify construction defects and generate construction defect reports; The construction management and correction module is used to transmit construction defect reports to the construction management platform and plan correction plans based on the construction defect reports; The verification and feedback module is used to compare the second three-dimensional model of the corrected building structure with the construction standard model to verify the correction effect after the construction defects are corrected.
4. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the construction error detection method based on three-dimensional laser scanning described in claim 1 or 2 are implemented.
5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the construction error detection method based on three-dimensional laser scanning described in claim 1 or 2 are implemented.
