Precast concrete dimension detection method and system

Abstract

The invention relates to the technical field of intelligent building industrial detection, and provides a concrete prefabricated part size detection method and system, and the method comprises the steps: obtaining the three-dimensional image data of a concrete prefabricated part, and carrying out the preprocessing of the obtained three-dimensional image data; extracting geometric features of the concrete prefabricated member from the preprocessed three-dimensional image data, and calculating the actual size of the concrete prefabricated member according to the extracted geometric features; and comparing the actual size of the concrete prefabricated part with a preset size, and generating a size detection result according to a comparison result. According to the precast concrete size detection method and system, the precision, efficiency and comprehensiveness of precast concrete size detection can be remarkably improved, the labor cost is reduced, and the data utilization value is enhanced.

Description

Technical Field

[0001] This application relates to the technical field of intelligent building industrialization detection, and particularly relates to a method and system for detecting the dimensions of concrete precast components. Background Technique

[0002] As an important building component, concrete precast components are widely used in modern building construction. Their dimensional accuracy not only directly affects the assembly accuracy and overall stability of the building structure, but also is related to construction efficiency and project quality. Therefore, high-precision dimensional detection of concrete precast components is a key link to ensure building quality. Traditional manual detection methods rely on measuring tools and manual operations, and have problems such as low efficiency, unstable accuracy, and large subjective errors. With the development of building industrialization and intelligent manufacturing, higher requirements are put forward for the automation, high precision, and high efficiency of the dimensional detection of concrete precast components.

[0003] In recent years, with the development of computer vision and three-dimensional scanning technology, automated detection of concrete precast components has been studied. The lack of detection accuracy is the main reason hindering the practical application of the automated detection of concrete precast components. Specifically, the existing automated detection methods for concrete precast components have the following deficiencies: 1. Insufficient ability to collect dimensional parameters, and only a few basic dimensional parameters of concrete precast components with simple structures can be collected. In particular, for the dimensional parameters of concrete precast components with complex structures, the collection of dimensional parameters cannot be achieved. 2. Insufficient ability to reduce noise in dimensional parameters, and it is difficult to remove the noise and background interference of dimensional parameters, resulting in inaccurate subsequent feature extraction. 3. Insufficient feature extraction ability, and it is difficult to set an effective feature extraction method according to dimensional parameters, and the collected dimensional parameters cannot be constructed into accurate three-dimensional contour features.

[0004] Chinese Patent with Publication No. CN111590746A discloses a detection method and system during the production process of concrete precast components, which inputs and converts the drawings of concrete precast components into images for storage to form an image database, and preset thresholds; performs real-time scanning on the formwork on the production line to obtain target image information and display it; compares the data features of the target image with the standard image, and judges whether it meets the design requirements through the preset threshold, and gives corresponding alarms. The technical solution of this patent relies on two-dimensional image comparison, detects two-dimensional data features such as the quantity, dimensions, and relative positions of molds, steel bars, and embedded parts during the production process, and does not involve the detection of the overall three-dimensional dimensions of precast components. Restricted by the expression ability of two-dimensional images, it is difficult to accurately extract and compare the features of precast components with complex shapes, and it is only applicable to the detection of processes such as mold assembly, steel bar installation, and embedded part installation during the production process of concrete precast components.

[0005] Therefore, how to provide a method for high-precision detection of the dimensions of concrete precast components to meet the strict requirements of modern building industrialization for the dimension detection of precast components has become a technical problem to be solved urgently.

[0006] Application Content In view of this, in order to overcome the deficiencies of the prior art, the present application aims to provide a method and system for detecting the dimensions of concrete precast components.

[0007] According to the first aspect of the present application, there is provided a method for detecting the dimensions of concrete precast components, the method comprising: Step S201: Obtain three-dimensional image data of the concrete precast component, and perform preprocessing on the obtained three-dimensional image data; Step S202: Extract geometric features of the concrete precast component from the preprocessed three-dimensional image data, and calculate the actual dimensions of the concrete precast component according to the extracted geometric features; Step S203: Compare the actual dimensions of the concrete precast component with the preset dimensions, and generate a dimension detection result according to the comparison result.

[0008] Optionally, in the method for detecting the dimensions of concrete precast components of the present application, step S201 includes: obtaining three-dimensional point cloud data of the concrete precast component, and performing noise reduction processing on the obtained three-dimensional point cloud data.

[0009] Optionally, in the method for detecting the dimensions of concrete precast components of the present application, in step S201, perform structured light scanning on the concrete precast component, and by projecting a coded grating and capturing the deformation of the reflected grating, reconstruct the surface three-dimensional point cloud data of the concrete precast component.

[0010] Optionally, in the method for detecting the dimensions of concrete precast components of the present application, in step S201, use a voxel grid-based downsampling algorithm to remove discrete points in the three-dimensional point cloud data, and use a region growing algorithm to remove background interference point cloud data to extract the main body point cloud data of the concrete precast component.

[0011] Optionally, in the method for detecting the dimensions of concrete precast components of the present application, step S202 includes: Calculate the normal vector of the three-dimensional image of the concrete precast component, and perform plane fitting on the three-dimensional image data of the concrete precast component according to the calculated normal vector to obtain edge contour features; Convert the extracted edge contour features into a continuous three-dimensional geometric model, and perform fitting optimization on the surface of the three-dimensional geometric model through surface fitting to obtain the actual dimensions of the concrete precast component.

[0012] Optionally, in the method for detecting the size of a concrete precast member of the present application, in step S202, the eigenvalues and eigenvectors of the covariance matrix of the three-dimensional point cloud data of the concrete precast member are calculated, the main direction and the secondary direction of the concrete precast member are determined according to the calculated eigenvalues and eigenvectors, and plane fitting is performed on the three-dimensional point cloud data according to the main direction and the secondary direction of the concrete precast member to obtain edge contour features.

[0013] Optionally, in the method for detecting the size of a concrete precast member of the present application, in step S202, the extracted edge contour features are converted into a continuous three-dimensional geometric model through a Poisson reconstruction algorithm, and the normal vectors and vertex positions of the three-dimensional point cloud data are optimized through surface fitting to obtain the actual size of the concrete precast member, where the actual size includes length, width, height, diagonal length, and flatness and perpendicularity of each plane.

[0014] Optionally, in the method for detecting the size of a concrete precast member of the present application, step S203 includes: calculating the deviation between the actual size and the preset size of the concrete precast member by constructing an error matrix, comparing the calculated deviation with a preset tolerance threshold, and when the deviation is not greater than the preset tolerance threshold, determining that the concrete precast member passes the size detection, and when the deviation is greater than the preset tolerance threshold, determining that the concrete precast member fails the size detection.

[0015] Optionally, in the method for detecting the size of a concrete precast member of the present application, in step S203, when it is determined that the concrete precast member fails the size detection, an alarm is issued through an acoustic-optic alarm device, an alarm log is generated, and the generated alarm log is stored in a database that supports remote access and data analysis, where the alarm log includes three-dimensional image data, geometric features, and size deviation information of the concrete precast member that fails the size detection.

[0016] According to a second aspect of the present application, there is provided a system for detecting the size of a concrete precast member, the system including a detection server, and the detection server includes: An image data acquisition module, configured to acquire three-dimensional image data of a concrete precast member and preprocess the acquired three-dimensional image data; A size calculation module, configured to extract geometric features of a concrete precast member from the preprocessed three-dimensional image data and calculate the actual size of the concrete precast member according to the extracted geometric features; A detection result generation module, configured to compare the actual size of the concrete precast member with a preset size and generate a size detection result according to the comparison result.

[0017] The method and system for detecting the size of a concrete precast member according to an embodiment of the present application have the following beneficial technical effects: 1. Improve detection accuracy: This application obtains high-precision three-dimensional point cloud data through structured light three-dimensional scanning, and uses data preprocessing algorithms to remove noise and background interference, ensuring the accuracy and integrity of the point cloud data. In addition, based on feature extraction and geometric modeling methods, the geometric features of precast concrete components are accurately extracted and a high-precision three-dimensional geometric model is generated, thus significantly improving the accuracy of dimension detection.

[0018] 2. Enhance detection efficiency: Through three-dimensional image data acquisition, data preprocessing, geometric feature extraction and calculation, and dimension comparison, manual intervention is avoided. At the same time, optimized algorithm design ensures the high efficiency of data processing, and can complete the comprehensive dimension detection of complex-shaped precast concrete components in a short time, significantly enhancing the detection efficiency and meeting the real-time detection needs of large-scale industrial production.

[0019] 3. Strengthen detection comprehensiveness: This application can not only detect the basic dimension parameters of precast concrete components (such as length, width, height), but also calculate key dimension indicators such as diagonal length, flatness and perpendicularity of each surface, realizing a comprehensive evaluation of the dimensions of precast components. In addition, through high-precision comparison with preset dimension standards, it can accurately judge whether the precast components meet the quality requirements, avoiding potential quality hazards caused by incomplete detection items in traditional methods.

[0020] 4. Achieve intelligent feedback: When detecting precast components with unqualified dimensions, this application can automatically issue an alarm and record detailed dimension deviation information, including specific dimension parameters, deviation values, and the three-dimensional geometric model of the unqualified part. These information will be stored in the database for subsequent quality analysis and formulation of improvement measures, thus realizing intelligent feedback in the detection process and improving the quality control level of the production process.

[0021] 5. Realize the detection of complex shapes: It can effectively handle precast concrete components with complex shapes. No matter how complex the shape of the precast component is, high-precision dimension detection can be achieved through accurate point cloud data processing and feature extraction. It is not only applicable to precast components with standard shapes, but also can meet the detection needs of diverse and complex components in modern architecture.

[0022] 6. Reduce labor costs: By realizing the automation and intelligence of the detection process, the dependence on manual measurement is reduced, thus significantly reducing labor costs.

[0023] 7. Improve the value of data utilization: Store the 3D geometric model and dimensional deviation information generated during the detection process in the database, providing rich data support for subsequent quality analysis, process optimization, and intelligent production. Through in-depth mining and analysis of these data, users can better grasp the quality fluctuations during the production process, adjust the production process in a timely manner, improve product quality and production efficiency. At the same time, the high-precision detection results reduce the defective rate caused by human errors, further reducing production costs and improving economic benefits. Description of the Drawings

[0024] To more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0025] Figure 1 It is an architecture example diagram of a concrete precast component size detection system according to an embodiment of the present application; Figure 2 It is an architecture example diagram of a detection server of a concrete precast component size detection system according to an embodiment of the present application; Figure 3 It is an execution flow example diagram of a concrete precast component size detection method according to an embodiment of the present application; Figure 4 It is another execution flow example diagram of a concrete precast component size detection method according to an embodiment of the present application. Detailed Embodiments

[0026] The embodiments of the present application will be described in detail below with reference to the drawings.

[0027] It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other; and, based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present disclosure.

[0028] Note that the following description pertains to various aspects of embodiments within the scope of the appended claims. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on this disclosure, those skilled in the art should understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of the aspects set forth herein can be used to implement a device and / or practice a method. Additionally, this device and / or method can be implemented using other structures and / or functionality in addition to one or more of the aspects described herein.

[0029] Figure 1 FIG. is an architecture example diagram of a concrete precast component size detection system according to an embodiment of the present application, as Figure 1 shown, the system may include a detection server 101, a communication network 102, and / or one or more detection clients 103, Figure 1 exemplified as multiple detection clients 103 in the figure.

[0030] The detection server 101 can be any suitable server for storing information, data, programs, and / or any other appropriate type of content. In some embodiments, the detection server 101 can perform appropriate functions. For example, in some embodiments, the detection server 101 can be used to detect the size of concrete precast components. As an optional example, in some embodiments, the detection server 101 can be used to: obtain three-dimensional image data of the concrete precast component, preprocess the obtained three-dimensional image data; extract geometric features of the concrete precast component from the preprocessed three-dimensional image data, calculate the actual size of the concrete precast component according to the extracted geometric features; compare the actual size of the concrete precast component with a preset size, and generate a size detection result according to the comparison result.

[0031] Figure 2 FIG. is an architecture example diagram of the detection server of a concrete precast component size detection system according to an embodiment of the present application, as Figure 2 shown, in this embodiment, the detection server includes: An image data acquisition module, configured to obtain three-dimensional image data of the concrete precast component and preprocess the obtained three-dimensional image data; A size calculation module, configured to extract geometric features of the concrete precast component from the preprocessed three-dimensional image data and calculate the actual size of the concrete precast component according to the extracted geometric features; A detection result generation module, configured to compare the actual size of the concrete precast component with a preset size and generate a size detection result according to the comparison result.

[0032] As another example, in some embodiments, the detection server 101 can, according to the request of the detection client 103, send the concrete precast component size detection method to the detection client 103 for the user to use.

[0033] As an alternative example, in some embodiments, the detection client 103 is used to provide a visual detection interface, which is used to receive the user's selection input operation for detecting the size of the concrete precast component, and, in response to the selection input operation, obtain from the detection server 101 the detection interface corresponding to the option selected by the selection input operation and display the detection interface, and at least the information for detecting the size of the concrete precast component and the operation options for the information for detecting the size of the concrete precast component are displayed in the detection interface.

[0034] In some embodiments, the communication network 102 can be any suitable combination of one or more wired and / or wireless networks. For example, the communication network 102 can include any one or more of the following: the Internet, an intranet, a wide area network (WAN), a local area network (LAN), a wireless network, a digital subscriber line (DSL) network, a frame relay network, an asynchronous transfer mode (ATM) network, a virtual private network (VPN), and / or any other suitable communication network. The detection client 103 can be connected to the communication network 102 through one or more communication links (for example, the communication link 104), and the communication network 102 can be linked to the detection server 101 through one or more communication links (for example, the communication link 105). The communication link can be any communication link suitable for transmitting data between the detection client 103 and the detection server 101, such as a network link, a dial-up link, a wireless link, a hard-wired link, any other suitable communication link, or any suitable combination of such links.

[0035] The detection client 103 can include any one or more clients that present an interface related to detecting the size of the concrete precast component in a suitable form for the user to use and operate. In some embodiments, the detection client 103 can include any suitable type of device. For example, in some embodiments, the detection client 103 can include a mobile device, a tablet computer, a laptop computer, a desktop computer, and / or any other suitable type of client device.

[0036] Although the detection server 101 is illustrated as one device, in some embodiments, any appropriate number of devices can be used to perform the functions performed by the detection server 101. For example, in some embodiments, multiple devices can be used to implement the functions performed by the detection server 101. Alternatively, the functions of the detection server 101 can be implemented using cloud services.

[0037] Based on the above system, an embodiment of the present application provides a method for detecting the dimensions of concrete precast components, which will be described below through the following embodiments.

[0038] Figure 3 FIG. is an example flowchart of the execution of a method for detecting the dimensions of concrete precast components according to an embodiment of the present application. The method for detecting the dimensions of concrete precast components in this embodiment can be executed on a detection server, such as Figure 3 shown, the method for detecting the dimensions of concrete precast components includes the following steps: Step S201: Obtain the three-dimensional image data of the concrete precast component, and preprocess the obtained three-dimensional image data.

[0039] As an optional example, in this embodiment, the three-dimensional point cloud data of the concrete precast component is obtained, and the obtained three-dimensional point cloud data is denoised. For example, in this embodiment, structured light scanning is performed on the concrete precast component, and by projecting a coded grating and capturing the deformation of the reflected grating, the three-dimensional point cloud data of the surface of the concrete precast component is reconstructed. In practical applications, the structured light scanning accuracy in this embodiment can be specifically set according to the application scenario. For example, the structured light scanning accuracy is set to not less than 0.01 mm. In this embodiment, the scanning range covers the entire surface of the concrete precast component.

[0040] The following will detail the structured light scanning and the acquisition of three-dimensional point cloud data in this embodiment in a specific scenario.

[0041] In this scenario, a batch of concrete precast components (precast slabs) need to be dimensionally inspected. The size of the precast slab is 2 m × 1 m × 0.2 m, and there may be minor manufacturing errors or defects on the surface. To ensure that the dimensional accuracy of the precast slab meets the design requirements, the method for detecting the dimensions of concrete precast components according to the embodiment of the present application is used for inspection, and it is implemented according to the following steps: A structured light 3D scanner equipped with a high-resolution camera and a projector is used. The projector is used to project a coded grating pattern, and the camera is used to capture the deformed image of the reflected grating. The accuracy of the scanner is set to 0.1 mm, which can meet the requirements for precast slab size inspection.

[0042] Place the concrete precast slab to be inspected on the workbench of the structured light 3D scanner, ensuring that its surface is flat and unobstructed. The projector of the scanner projects a series of coded grating patterns onto the surface of the precast slab. These grating patterns are specially designed stripe patterns, and their deformation on the surface of the precast slab can reflect the three-dimensional shape of the surface. As the surface of the precast slab undulates and changes in shape, the reflected grating pattern will deform accordingly. The camera of the structured light 3D scanner captures these deformed grating patterns and converts them into digital image data. For example, the protrusions on the surface of the precast slab will cause the grating stripes to bend, while the depressions will cause the stripe spacing to become larger or smaller.

[0043] The structured light 3D scanner obtains the 3D information of the precast slab surface by projecting and capturing grating patterns from multiple angles, acquiring 3D information of the precast slab surface from different directions. The scan at each angle can generate an image containing the grating deformation information. By performing multiple scans (e.g., from different angles such as the front, side, and inclined planes), comprehensive data of the precast slab surface can be obtained. For example, the structured light 3D scanner acquires 10 images in each direction, for a total of 30 images. These images contain the grating deformation information of the precast slab surface in different directions, providing a rich data basis for subsequent 3D reconstruction.

[0044] The collected image data is transmitted to a computer processing system for the reconstruction of 3D point cloud data. First, each image is preprocessed, including noise removal, camera distortion correction, etc. Then, through an algorithm to analyze the deformation of the grating pattern in the image, the 3D coordinates of each point on the precast slab surface are calculated. Specifically, the phase change of the grating stripes is used to determine the depth information of each point. For example, when the grating stripes bend on the precast slab surface, by calculating the phase offset of the stripes, combined with the known grating pattern and camera parameters, the depth coordinate of this point is deduced, and the 3D shape of the precast slab surface is reconstructed point by point, generating a dense 3D point cloud model containing a large number of point coordinates on the precast slab surface. For example, for a precast slab with dimensions of 2 m × 1 m × 0.2 m, the point cloud data contains millions of points, and the coordinate accuracy of each point reaches 0.1 mm.

[0045] The generated 3D point cloud data can be used for various subsequent processing and analysis. For example, by calculating the bounding box of the point cloud data, the actual size of the precast slab can be quickly obtained. At the same time, the point cloud data can also be used to detect defects on the precast slab surface, such as cracks, holes, or uneven areas. In addition, the point cloud data can be further converted into a 3D geometric model for more complex analysis, such as comparison with design drawings, dimensional deviation calculation, etc.

[0046] As an optional example, after obtaining the 3D point cloud data of the concrete precast member in this embodiment, this embodiment uses a voxel grid-based downsampling algorithm to remove the discrete points in the 3D point cloud data, and uses a region growing algorithm to remove the background interference point cloud data, extracting the main body point cloud data of the concrete precast member. In this embodiment, the resolution of the voxel grid is dynamically adjusted according to the size and accuracy of the concrete precast member to ensure the efficiency and accuracy of data processing.

[0047] When implementing the method of this application embodiment, the 3D point cloud data of the obtained concrete precast member can also be subjected to filtering processing and normalization processing to remove noise and outliers.

[0048] Step S202: Extract the geometric features of the precast concrete member from the preprocessed three-dimensional image data, and calculate the actual size of the precast concrete member according to the extracted geometric features.

[0049] As an optional example, in this embodiment, calculate the normal vector of the three-dimensional image of the precast concrete member, and perform plane fitting on the three-dimensional image data of the precast concrete member according to the calculated normal vector to obtain the edge contour feature. In practical applications, in this embodiment, calculate the eigenvalues and eigenvectors of the covariance matrix of the three-dimensional point cloud data of the precast concrete member, determine the main direction and the secondary direction of the precast concrete member according to the calculated eigenvalues and eigenvectors, and perform plane fitting on the three-dimensional point cloud data according to the main direction and the secondary direction of the precast concrete member to obtain the edge contour feature. For example, in this embodiment, the principal component analysis (PCA) algorithm can be used to perform plane fitting on the three-dimensional point cloud data. The principal component analysis (Principal Component Analysis, PCA) algorithm is used to project multi-dimensional data into a low-dimensional space through linear transformation, while retaining the variance information of the original data, reducing the complexity of the data through dimensionality reduction technology, and at the same time extracting the most important features (i.e., the principal components) in the data.

[0050] The following will give a detailed description of the processing of the three-dimensional point cloud data of the precast concrete member and the extraction of the edge contour feature in the present application in a specific scenario.

[0051] In this scenario, it is necessary to detect the sizes of a batch of precast concrete slabs. The size of the precast slab is 2m×1m×0.2m, and there may be minor manufacturing errors or defects on the surface. In order to ensure that the size accuracy of the precast slab meets the design requirements, the precast concrete member size detection method of the embodiment of the present application is used for detection, and it is implemented according to the following steps: 1. Covariance matrix calculation After preprocessing the three-dimensional point cloud data, including removing noise and background interference, calculate the covariance matrix of the point cloud data. The covariance matrix describes the correlation between the features in the point cloud data, and the calculation formula is:

[0052] Among them, is the preprocessed point cloud data matrix, is its transpose, is the number of points in the point cloud data.

[0053] Suppose the point cloud data matrix has a dimension of , where is the number of points, and 3 represents the x, y, and z coordinates of each point. Through calculation, a 3×3 covariance matrix ∑ is obtained.

[0054] 2. Calculation of Eigenvalues and Eigenvectors Solve the eigenvalues and eigenvectors of the covariance matrix ∑. The eigenvalues represent the variance magnitudes of each principal component, and the eigenvectors represent the directions of the principal components. By solving the characteristic equation: det(∑ - λI) = 0, three eigenvalues λ1, λ 2， λ 3， and the corresponding eigenvectors v1, v 2， v 3， Suppose the calculated eigenvalues are λ1 = 0.001, λ2 = 0.0005, λ3 = 0.0001, and the corresponding eigenvectors are v1 = [0.6, 0.8, 0], v2 = [-0.8, 0.6, 0], v3 = [0, 0, 1]).

[0055] 3. Determination of the Main Direction and Secondary Direction Sort the eigenvectors according to the magnitudes of the eigenvalues, and select the eigenvectors corresponding to the two largest eigenvalues, namely v1 and v2, as the main direction and secondary direction of the precast concrete slab. In this example, the main direction is v1 = [0.6, 0.8, 0], and the secondary direction is v2 = [-0.8, 0.6, 0].

[0056] 4. Plane Fitting and Edge Contour Feature Extraction According to the determined main direction and secondary direction, perform plane fitting on the three-dimensional point cloud data. The purpose of plane fitting is to find a plane such that the projection of the point cloud data on this plane can retain the variance information of the original data as much as possible. The plane equation can be expressed as: ax + by + cz = d where a, b, c are the normal vectors of the plane, which can be obtained by the cross product of the main direction and the secondary direction. In this example, the normal vector is v3 = [0, 0, 1], so the plane equation is simplified to: z = d.

[0057] By methods such as the least squares method, the value of d in the plane equation can be solved. Suppose the solved d = 0.1, then the plane equation is: z = 0.1.

[0058] Project the three-dimensional point cloud data onto this plane to obtain two-dimensional point cloud data. Then, by analyzing the distribution of the two-dimensional point cloud data, the edge contour features of the precast concrete slab can be extracted. For example, by finding the extreme points in the point cloud data or using edge detection algorithms, the four edge positions of the precast slab can be determined.

[0059] After obtaining the edge contour features of the concrete precast member, in this embodiment, the extracted edge contour features are converted into a continuous three-dimensional geometric model, and the surface of the three-dimensional geometric model is fitted and optimized through surface fitting to obtain the actual size of the concrete precast member. For example, in this embodiment, the extracted edge contour features are converted into a continuous three-dimensional geometric model through the Poisson reconstruction algorithm, and the normal vector and vertex position of the three-dimensional point cloud data are optimized through surface fitting to obtain the actual size of the concrete precast member. In practical applications, by optimizing the normal vector and vertex position of the three-dimensional point cloud data, the surface error of the three-dimensional geometric model is controlled. For example, the surface error of the three-dimensional geometric model is controlled to be less than 0.05 mm. In this embodiment, the actual size includes length, width, height, diagonal length, and flatness and perpendicularity of each plane. For example, the diagonal length is calculated through the vertex coordinates of the three-dimensional geometric model.

[0060] In the following, in another specific scenario, the processing of the three-dimensional point cloud data of the concrete precast member and the extraction of the edge contour features in this application are described in detail.

[0061] In this scenario, it is necessary to detect the sizes of a batch of concrete precast members with complex and irregular shapes. The shapes of these precast members include various complex structures such as polygons, curved surfaces, grooves, and protrusions, and there may be minor manufacturing errors or defects on the surface. The concrete precast members in this scenario include a main body part, a top structure, and a side structure. The main body part is an irregular polygon base with dimensions of approximately 2 m × 1.5 m × 0.5 m; the top structure is a protrusion with a curved surface and a height of approximately 0.3 m; the side structure includes multiple grooves and protrusions with complex shapes.

[0062] To ensure that the size accuracy of the precast member meets the design requirements, the size detection method of the concrete precast member in the embodiment of this application is used for detection, and it is implemented according to the following steps: 1. Covariance matrix calculation After preprocessing the three-dimensional point cloud data, including removing noise and background interference, the covariance matrix of the point cloud data is calculated. The covariance matrix describes the correlation between the features in the point cloud data, and the calculation formula is:

[0063] Among them, is the preprocessed point cloud data matrix, is its transpose, is the number of points in the point cloud data.

[0064] Assume that the point cloud data matrix has a dimension of where, is the number of points, and 3 represents the x, y, and z coordinates of each point. Through calculation, a 3×3 covariance matrix ∑ is obtained.

[0065] 2. Eigenvalue and eigenvector calculation Solve the eigenvalues and eigenvectors of the covariance matrix ∑. The eigenvalues represent the variance magnitudes of each principal component, and the eigenvectors represent the directions of the principal components. By solving the characteristic equation: det(∑ - λI) = 0, three eigenvalues λ1, λ 2， λ 3， and the corresponding eigenvectors v1, v 2， v 3， Suppose the calculated eigenvalues are λ1 = 0.002, λ2 = 0.0001, λ3 = 0.0005, and the corresponding eigenvectors are v1 = [0.7, 0.5, 0.5], v2 = [-0.5, 0.8, 0], v3 = [0, 0, 1]).

[0066] 3. Determination of the main direction and secondary direction Sort the eigenvectors according to the magnitudes of the eigenvalues, and select the eigenvectors corresponding to the two largest eigenvalues, namely v1 and v2, as the main direction and secondary direction of the precast concrete component. In this example, the main direction is v1 = [0.7, 0.5, 0.5], and the secondary direction is v2 = [-0.5, 0.8, 0].

[0067] 4. Plane fitting and edge contour feature extraction According to the determined main direction and secondary direction, perform plane fitting on the three-dimensional point cloud data. The purpose of plane fitting is to find a plane such that the projection of the point cloud data on this plane retains the variance information of the original data as much as possible. The plane equation is expressed as: ax + by + cz = d where a, b, c are the normal vectors of the plane, which can be obtained by the cross product of the main direction and the secondary direction. In this example, the normal vector is v3 = [0, 0, 1], so the plane equation is simplified to: z = d.

[0068] By methods such as the least squares method, the value of d in the plane equation can be solved. Suppose the solved d = 0.1, then the plane equation is: z = 0.1.

[0069] Project the three-dimensional point cloud data onto this plane to obtain two-dimensional point cloud data. Then, by analyzing the distribution of the two-dimensional point cloud data, the edge contour features of the precast concrete component can be extracted. For example, by finding the extreme points in the point cloud data or using edge detection algorithms, the edge positions of the precast component can be determined.

[0070] In this scenario, it is also necessary to perform plane fitting and edge extraction on multiple local regions respectively. The edge contour of the base is extracted through plane fitting. The edge contour of the base of the main body part is a polygon, and the vertex coordinates are (0, 0), (2, 0), (2, 1.5), and (0, 1.5).

[0071] The surface boundary of the top structure is extracted through surface fitting. The boundary obtained by surface fitting is an ellipse, and the center coordinates are (1, 0.75, 0.5).

[0072] The edges of the grooves and protrusions of the side structure are extracted through local plane fitting, which are multiple line segments, and the coordinates are (0.5, 0.2, 0.1), (0.5, 0.8, 0.1), (1.5, 0.2, 0.1), and (1.5, 0.8, 0.1).

[0073] Through the above steps, the edge contour features of the concrete precast member with a complex shape can be accurately extracted, providing an accurate and complete data basis for subsequent dimension calculation and quality assessment.

[0074] Step S203: Compare the actual size of the concrete precast member with the preset size, and generate a size detection result according to the comparison result.

[0075] Figure 4 Another execution flow example diagram of a method for detecting the size of a concrete precast member according to an embodiment of the present application is shown in Figure 4 As shown, as an optional example, in this embodiment, the deviation between the actual size and the preset size of the concrete precast member is calculated by constructing an error matrix, and the calculated deviation is compared with a preset tolerance threshold. When the deviation is not greater than the preset tolerance threshold, it is determined that the concrete precast member passes the size detection. When the deviation is greater than the preset tolerance threshold, it is determined that the concrete precast member fails the size detection.

[0076] It should be noted that in this embodiment, when it is determined that the concrete precast member fails the size detection, an alarm is issued through an acoustic-optic alarm device, an alarm log is generated, and the generated alarm log is stored in a database that supports remote access and data analysis. The alarm log includes three-dimensional image data, geometric features, and size deviation information of the concrete precast member that fails the size detection.

[0077] The following will provide a detailed description of the three-dimensional point cloud data processing and edge contour feature extraction of the concrete precast member in the present application in a specific scenario.

[0078] In this scenario, it is necessary to detect the dimensions of a batch of precast concrete slabs. The dimensions of the precast slabs are 5 meters × 0.5 meters × 0.8 meters, and there may be minor manufacturing errors or defects on the surface. To ensure that the dimensional accuracy of the precast slabs meets the design requirements, the concrete precast component dimension detection method of this application embodiment is used for detection, and it is implemented according to the following steps: 1. Normal vector estimation Calculate the local normal vector for each point cloud data point. By calculating the covariance matrix of the neighborhood points and solving its eigenvalues and eigenvectors, the normal vector of each point is obtained. These normal vectors will be used as one of the inputs for Poisson reconstruction.

[0079] 2. Poisson reconstruction Use the Poisson reconstruction algorithm to construct a continuous three-dimensional geometric model based on the point cloud data and its normal vectors. The Poisson reconstruction algorithm optimizes the vertex positions and normal vectors of the point cloud data by solving the Poisson equation to generate a smooth three-dimensional surface. For example, for a precast beam with a design size of 5 meters × 0.5 meters × 0.8 meters, the Poisson reconstruction algorithm can generate a three-dimensional geometric model with high precision, and the surface error of the model does not exceed 0.05 millimeters.

[0080] 3. Surface fitting optimization To further improve the accuracy of the three-dimensional geometric model, the surface of the model is optimized by surface fitting in this embodiment. The specific steps are as follows: 1). Perform surface fitting on each surface of the generated three-dimensional geometric model. For example, for the top and bottom surfaces of the precast beam, plane fitting can be used; for the side surfaces, cylindrical surface fitting can be used. By optimization methods such as the least squares method, adjust the vertex positions on the model surface to make the model surface smoother and conform to the actual shape. 2). During the surface fitting process, simultaneously optimize the normal vectors and vertex positions of the point cloud data. Through iterative optimization, ensure that the normal vectors of the model surface are consistent with the normal vectors of the actual surface, and at the same time reduce the error of the vertex positions. For example, during the optimization process, the adjustment accuracy of the vertex positions reaches 0.01 millimeters, and the adjustment accuracy of the normal vectors reaches 0.1 degrees.

[0081] 4. Dimension calculation After completing the surface fitting optimization, obtain the actual dimensions of the precast concrete beam by calculating the geometric parameters of the three-dimensional geometric model. The specific steps are as follows: 1). Calculate the bounding box: Calculate the bounding box of the three-dimensional geometric model to obtain the actual length, width, and height of the precast beam. The calculation formula for the bounding box is: Length = max(x) - min(x) Width = max(y) - min(y) Height = max(z) - min(z) Assume that in this scenario, the bounding box of the optimized three-dimensional geometric model is as follows: Length = 5.002 meters Width = 0.501 meters Height = 0.801 meters 2) Calculate the diagonal length To further verify the accuracy of the dimensions, calculate the diagonal length of the precast beam. The calculation formula for the diagonal length is:

[0082] Substitute the actual dimensions into the calculation to obtain: .

[0083] 5. Dimension comparison and result judgment Compare the calculated actual dimensions with the preset design dimensions to determine whether the precast beam meets the quality requirements.

[0084] Assume that the preset tolerance range is ±0.5 mm, then: Length deviation: 5.002 m - 5.000 m = 0.002 m Width deviation: 0.501 m - 0.500 m = 0.001 m Height deviation: 0.801 m - 0.800 m = 0.001 m All deviations are within the tolerance range, so it is determined that the dimensions of the precast beam are qualified.

[0085] According to the concrete precast component size detection method and system of the embodiments of the present application, the following beneficial technical effects are achieved: 1. Improve detection accuracy: The present application obtains high-precision three-dimensional point cloud data through structured light three-dimensional scanning, and uses data preprocessing algorithms to remove noise and background interference to ensure the accuracy and integrity of the point cloud data; in addition, based on feature extraction and geometric modeling methods, geometric features of concrete precast components are accurately extracted and a high-precision three-dimensional geometric model is generated, thereby significantly improving the accuracy of size detection.

[0086] 2. Improve detection efficiency: Through three-dimensional image data acquisition, data preprocessing, geometric feature extraction and calculation, and dimension comparison, manual intervention is avoided; at the same time, the optimized algorithm design ensures the high efficiency of data processing, and can complete the comprehensive size detection of complex-shaped concrete precast components in a short time, significantly improving the detection efficiency and meeting the real-time detection requirements of large-scale industrial production.

[0087] 3. Enhance the comprehensiveness of detection: This application can not only detect the basic dimensional parameters of precast concrete components (such as length, width, and height), but also calculate key dimensional indicators such as diagonal length, flatness, and perpendicularity of each surface, achieving a comprehensive assessment of the dimensions of precast components. In addition, through a high-precision comparison with the preset dimensional standards, it can accurately determine whether the precast components meet the quality requirements, avoiding potential quality hazards caused by incomplete detection items in traditional methods.

[0088] 4. Achieve intelligent feedback: When detecting precast components with unqualified dimensions, this application can automatically issue an alarm and record detailed dimensional deviation information, including specific dimensional parameters, deviation values, and the 3D geometric model of the unqualified part. This information will be stored in the database for subsequent quality analysis and formulation of improvement measures, thereby achieving intelligent feedback in the detection process and improving the quality control level of the production process.

[0089] 5. Achieve the detection of complex shapes: It can effectively handle precast concrete components with complex shapes. Regardless of how complex the shape of the precast component is, high-precision dimensional detection can be achieved through precise point cloud data processing and feature extraction. It is not only applicable to precast components with standard shapes but also meets the detection requirements of diverse and complex components in modern architecture.

[0090] 6. Reduce labor costs: By automating and intelligentizing the detection process, the dependence on manual measurement is reduced, thereby significantly reducing labor costs.

[0091] 7. Enhance the value of data utilization: Store the 3D geometric model and dimensional deviation information generated during the detection process in the database, providing rich data support for subsequent quality analysis, process optimization, and intelligent production. Through in-depth mining and analysis of this data, users can better grasp the quality fluctuations in the production process, timely adjust the production process, improve product quality and production efficiency. At the same time, the high-precision detection results reduce the defective rate caused by human errors, further reducing production costs and improving economic benefits.

[0092] Through the description of the above implementation manners, those skilled in the art can clearly understand that each implementation manner can be realized by means of software plus a necessary general hardware platform, and of course, it can also be realized by hardware. Based on such an understanding, the above technical solutions, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disc, etc., including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiments.

[0093] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present application should be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for detecting the size of precast concrete components, characterized in that, The method includes: Step S201: Obtain the three-dimensional image data of the concrete precast member, and preprocess the obtained three-dimensional image data; Step S202: Extract the geometric features of the concrete precast member from the preprocessed three-dimensional image data, and calculate the actual size of the concrete precast member according to the extracted geometric features; Step S203: Compare the actual size of the concrete precast member with the preset size, and generate a size detection result according to the comparison result.

2. The method for detecting the size of precast concrete components according to claim 1, characterized in that, Step S201 includes: obtaining the three-dimensional point cloud data of the concrete precast member, and performing noise reduction processing on the obtained three-dimensional point cloud data.

3. The method for detecting the size of the precast concrete member according to claim 2, wherein, In step S201, perform structured light scanning on the concrete precast member, and reconstruct the surface three-dimensional point cloud data of the concrete precast member by projecting a coded grating and capturing the deformation of the reflected grating.

4. The method for detecting the size of precast concrete components according to claim 2, characterized in that, In step S201, use a voxel grid-based downsampling algorithm to remove discrete points in the three-dimensional point cloud data, and use a region growing algorithm to remove background interference point cloud data to extract the main body point cloud data of the concrete precast member.

5. The method for detecting the size of precast concrete components according to claim 1, characterized in that, Step S202 includes: Calculate the normal vector of the three-dimensional image of the concrete precast member, perform plane fitting on the three-dimensional image data of the concrete precast member according to the calculated normal vector, and obtain the edge contour feature; Convert the extracted edge contour feature into a continuous three-dimensional geometric model, perform surface fitting optimization on the surface of the three-dimensional geometric model through surface fitting, and obtain the actual size of the concrete precast member.

6. The method for detecting the size of precast concrete components according to claim 5, wherein, In step S202, calculate the eigenvalues and eigenvectors of the covariance matrix of the three-dimensional point cloud data of the concrete precast member, determine the main direction and secondary direction of the concrete precast member according to the calculated eigenvalues and eigenvectors, and perform plane fitting on the three-dimensional point cloud data according to the main direction and secondary direction of the concrete precast member to obtain the edge contour feature.

7. The method for detecting the size of precast concrete components according to claim 5, characterized in that, In step S202, convert the extracted edge contour feature into a continuous three-dimensional geometric model through the Poisson reconstruction algorithm, optimize the normal vector and vertex position of the three-dimensional point cloud data through surface fitting, and obtain the actual size of the concrete precast member, where the actual size includes length, width, height, diagonal length, and flatness and perpendicularity of each plane.

8. The method for detecting the size of precast concrete components according to claim 1, wherein, Step S203 includes: calculating the deviation between the actual size and the preset size of the concrete precast member by constructing an error matrix, comparing the calculated deviation with a preset tolerance threshold, when the deviation is not greater than the preset tolerance threshold, determining that the concrete precast member passes the size detection, and when the deviation is greater than the preset tolerance threshold, determining that the concrete precast member fails the size detection.

9. The method for detecting the size of precast concrete components according to claim 1, wherein In step S203, when it is determined that the concrete precast member fails the size detection, an alarm is issued through an acoustic-optic alarm device, an alarm log is generated, and the generated alarm log is stored in a database that supports remote access and data analysis. The alarm log includes the three-dimensional image data, geometric features, and size deviation information of the concrete precast member that fails the size detection.

10. A concrete precast component size detection system, characterized in that, The system includes a detection server, and the detection server includes: An image data acquisition module for obtaining the three-dimensional image data of the concrete precast member and preprocessing the obtained three-dimensional image data; A dimension calculation module, configured to extract geometric features of a precast concrete member from the preprocessed three-dimensional image data and calculate the actual dimensions of the precast concrete member according to the extracted geometric features; A detection result generation module, configured to compare the actual dimensions of the precast concrete member with preset dimensions and generate a dimension detection result according to the comparison result.

Images