Method, device and medium for detecting dimensional errors of steel structure components based on point cloud

Abstract

The present invention discloses a method, device and medium for detecting dimensional errors of steel structural components based on point clouds. The method comprises: performing point cloud scanning on the steel structural component to be measured to obtain point cloud data, and combining multiple preset marked points on the steel structural component to be measured to obtain the point cloud coordinate value of each preset marked point, and calculating the actual distance between any two preset marked points based on the point cloud coordinate values ​​of the multiple preset marked points; when there is a dimensional error between the actual distance and the design parameters of the steel structural component to be measured, constructing a point cloud model based on the point cloud data and matching it with the design model, obtaining deviation data of the steel structural component to be measured based on the matched difference model, and displaying the difference model and deviation data through a visual display platform. The present invention can provide a digital, intelligent and visual method for detecting dimensional errors of steel structural components, solving the problems of insufficient intelligence, large errors and low efficiency caused by existing manual measurement.

Description

Technical Field

[0001] The present invention relates to quality inspection and dimension monitoring of steel structure components, and in particular to a method, device and medium for dimension error detection of steel structure components based on point cloud. Background Art

[0002] Currently, the dimensional inspection and acceptance of steel structure components still relies primarily on manual methods, commonly employing traditional methods such as stringing and hanging wires, supplemented by measuring tools such as theodolites, levels, or total stations for dimensional inspection and positioning verification. In practice, this process often requires the coordination of multiple operators, with one person responsible for stringing and positioning, another taking readings and recording, and still others climbing onto the upper surface of the steel structure for measurement and calibration. This method not only requires high levels of coordination and operational experience, but also creates a complex working environment and presents significant safety risks. Furthermore, inspection results are often recorded manually, lacking automated data collection and digital storage. This makes it impossible to visualize and dynamically track the structural status, hindering subsequent data analysis and quality traceability. The overall process is cumbersome, inefficient, and significantly impacted by human factors. This limitation and potential safety hazards are particularly prominent in steel structure projects involving large spans, complex structures, or those working at height. Therefore, there is an urgent need for a new type of steel structure dimension detection and monitoring technology that is efficient, intelligent, highly automated, and has safety assurance capabilities to replace traditional manual measurement methods and improve the quality control level and operational safety during the construction process. Summary of the Invention

[0003] In order to overcome the shortcomings of the existing technology, one of the purposes of the present invention is to provide a point cloud-based steel structure component dimensional error detection method, which can solve the problems of low efficiency, cumbersome operation and low safety in the existing technology of steel structure component dimensional error detection.

[0004] The second purpose of the present invention is to provide a point cloud-based steel structure component dimensional error detection device, which can solve the problems of low efficiency, cumbersome operation and low safety in the existing technology when detecting the size of steel structure components.

[0005] A third object of the present invention is to provide a computer-readable storage medium that can solve the problems of low efficiency, complicated operation and low safety in the existing technology for steel structure component size detection.

[0006] One of the purposes of the present invention is achieved by the following technical solution:

[0007] A method for detecting dimensional errors of steel structure components based on point clouds, wherein the method comprises:

[0008] Point cloud acquisition step: performing point cloud scanning on the steel structure component to be measured to obtain point cloud data of the steel structure component to be measured;

[0009] Distance calculation step: obtaining the point cloud coordinate value of each preset marked point according to the plurality of preset marked points and the point cloud data on the steel structure member to be measured, and calculating the actual distance between any two preset marked points according to the point cloud coordinate values ​​of the plurality of preset marked points;

[0010] Determining step: determining whether the steel structure component to be measured has a dimensional error according to the actual distance between any two preset marked points and the design parameters of the steel structure component to be measured, and if so, executing the matching step;

[0011] Matching step: constructing a point cloud model of the steel structure component to be measured based on the point cloud data of the steel structure component to be measured, and matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured to obtain a difference model, and then obtaining deviation data of the point cloud model of the steel structure component to be measured relative to the design model of the steel structure component to be measured based on the difference model;

[0012] Display push step: displaying the difference model and deviation data to staff through a visual display platform.

[0013] Furthermore, the point cloud acquisition step specifically includes: first, marking multiple preset marking points on the steel structure component to be measured according to the characteristics of the steel structure component to be measured, and then performing point cloud scanning on the steel structure component to be measured by a three-dimensional laser scanning device to obtain point cloud data of the structural component to be measured.

[0014] Furthermore, the judgment step specifically includes: first obtaining the design parameters of the steel structure component to be measured and then deriving the standard distance between any two preset marked points of the steel structure component to be measured based on the design parameters of the steel structure component to be measured; then matching the actual distance between each two preset marked points with the corresponding standard distance, and then judging whether there is a dimensional error in the steel structure component to be measured based on the matching result; wherein, when the actual distance between any two preset marked points in the matching result does not match the corresponding standard distance, the steel structure component to be measured has a dimensional error.

[0015] Furthermore, matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured in the matching step specifically includes: matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured based on a spatial matching algorithm and a coordinate transformation algorithm.

[0016] Furthermore, in the matching step, matching the point cloud model of the steel structure component to be measured with the design drawing of the steel structure component to be measured to obtain a difference model specifically includes: obtaining the difference model when the sum of the squares of the distances between the points in the point cloud model of the steel structure component to be measured and the points in the design model of the steel structure component to be measured is minimized after the point cloud model of the steel structure component to be measured is scaled, translated, and rotated multiple times.

[0017] Furthermore, in the matching step, matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured to obtain a difference model further specifically includes:

[0018] Step 1: according to each preset marked point, find out the first coordinate value of each preset marked point from the point cloud model of the steel structure component to be measured;

[0019] Step 2: according to each preset marked point, find out the second coordinate value of each preset marked point from the design model of the steel structure component to be measured;

[0020] Step 3: Match the point cloud model and the design model after multiple scaling, rotation, and translation to ensure that the sum of the squares of the distances between the multiple first coordinate values ​​and the corresponding second coordinate values ​​is minimized; and when the sum of the squares of the distances between the multiple first coordinate values ​​and the corresponding second coordinate values ​​is minimized, a difference model is obtained based on the matching results of the point cloud model and the design model.

[0021] Furthermore, the display push step includes: obtaining the deviation direction and deviation amount of the steel structure component to be measured based on the deviation data, and displaying the difference model, deviation data and deviation amount to the staff through a visual display platform.

[0022] Furthermore, the display push step also includes: matching the corresponding adjustment scheme from the scheme matching model constructed in the system according to the type, deviation direction and deviation amount of the steel structure component to be tested, and pushing the corresponding adjustment scheme to the staff; through statistical history of the deviation direction, deviation amount and corresponding adjustment scheme of the same type of steel structure components, and combining the machine learning model to obtain the matching relationship between the deviation direction, deviation amount and adjustment scheme of the steel structure component to construct a scheme matching model for the corresponding type of steel structure component.

[0023] The second object of the present invention is achieved by adopting the following technical solution:

[0024] A point cloud-based steel structure component dimensional error detection device includes:

[0025] A point cloud acquisition module is used to perform point cloud scanning on the steel structure component to be measured to obtain point cloud data of the steel structure component to be measured;

[0026] a distance calculation module, configured to obtain a point cloud coordinate value of each preset marked point based on the plurality of preset marked points and the point cloud data on the steel structure member to be measured, and to calculate the actual distance between any two preset marked points based on the point cloud coordinate values ​​of the plurality of preset marked points;

[0027] a judgment module, configured to judge whether the steel structure component to be measured has a dimensional error based on the actual distance between any two preset marked points and the design parameters of the steel structure component to be measured, and if so, perform a matching step;

[0028] a matching module, configured to construct a point cloud model of the steel structure component to be measured based on the point cloud data of the steel structure component to be measured, and match the point cloud model of the steel structure component to be measured with a design model of the steel structure component to be measured to obtain a difference model, and further obtain deviation data of the point cloud model of the steel structure component to be measured relative to the design model of the steel structure component to be measured based on the difference model;

[0029] The display push module is used to display the difference model and deviation data to the staff through a visual display platform.

[0030] The third object of the present invention is achieved by adopting the following technical solution:

[0031] A computer-readable storage medium stores a steel structure component dimensional error detection program, which is a computer program. When the steel structure component dimensional error detection program is executed by a processor, it implements the steps of the point cloud-based steel structure component dimensional error detection method adopted as one of the purposes of the present invention.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The present invention first performs point cloud scanning on steel structural components to obtain the marked points of the steel structural components to obtain the dimensional data of the steel structural components, and then determines whether the steel structural components meet the requirements. Then, for steel structural components that do not meet the requirements, the point cloud model constructed based on the point cloud data is matched with the design model to further compare and judge the dimensional errors of the steel structural components. The detected differences are displayed to the staff through a visual display platform, making the detection results more intuitive. The present invention solves the problem of improving detection efficiency by matching the scanned actual model data with the design model data to determine whether the actual size of the steel structural components meets the design data, while also displaying the measurement results through a visual platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A flow chart of the method for detecting dimensional errors of steel structural components based on point clouds provided by the present invention;

[0035] Figure 2 for Figure 1 Flowchart of step S4 in FIG.

[0036] Figure 3 This is a dimensional schematic diagram obtained after matching the point cloud model provided by the present invention with the design model. DETAILED DESCRIPTION

[0037] The present invention will be further described below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the various embodiments or technical features described below can be arbitrarily combined to form new embodiments.

[0038] Example 1

[0039] The present invention uses the physical data of steel structure components and the design parameter data to match them to detect the dimensional errors of steel structure components, thus solving the problems of low measurement efficiency and large measurement errors in the existing technology when manual measurement of steel structure components is required. At the same time, it can also avoid the measurement personnel from climbing up and down, thus ensuring the safety of the measurement personnel. The dimensional error detection of steel structure components provided by the present invention can greatly improve the efficiency and accuracy of dimensional error detection of steel structure components. Specifically, Figure 1 As shown, the present invention provides a preferred embodiment, a method for detecting dimensional errors of steel structural components based on point clouds, comprising:

[0040] Step S1: performing point cloud scanning on the steel structure component to be measured to obtain point cloud data of the steel structure component to be measured.

[0041] Specifically, the steel structure component to be tested mentioned in this embodiment refers to a steel structure component that currently needs to be tested for dimensional error, and does not refer to a specific steel structure component.

[0042] This embodiment first uses three-dimensional laser scanning technology to perform point cloud scanning on the steel structure component to be measured to obtain point cloud data. The actual size data of the steel structure component to be measured can be obtained based on the scanned point cloud data. By matching and comparing the actual size data of the steel structure component to be measured with the design size data, the size error detection of the steel structure component to be measured can be realized, solving the problems of poor efficiency and large errors existing in manual measurement, significantly improving the accuracy and efficiency of detection and calibration, reducing human intervention and operational difficulty, and solving the problems of insufficient measurement accuracy and inability to mark the size of steel structure components in the existing technology.

[0043] More specifically, in this embodiment, when scanning a steel structure component to be tested, a point cloud scan can be performed using a 3D laser scanning device. The 3D laser scanning device can be either handheld or fixed. When the 3D laser scanning device is handheld, a worker can use the 3D laser scanning device to scan the exterior of the steel structure component to obtain point cloud data. When the 3D laser scanning device is fixed, multiple 3D laser scanning devices can be used to simultaneously scan different surfaces of the steel structure component, and the resulting point cloud data can be merged to obtain the point cloud data of the steel structure component.

[0044] Furthermore, after acquiring the point cloud data of the steel structure component, this embodiment also preprocesses the point cloud data. Specifically, preprocessing includes removing interference points and invalid points. For example, interference points or invalid points may be present in the scanned point cloud data due to factors such as light or other obstacles. Therefore, it is necessary to remove interference points and invalid points from the point cloud data of the steel structure component to improve the accuracy of subsequent data processing.

[0045] More preferably, this embodiment further includes, before performing point cloud scanning on the steel structure component, marking a plurality of preset marking points on the steel structure component to be measured according to the characteristics of the steel structure component to be measured. In this way, the point cloud coordinate value of each preset marking point can be obtained after performing point cloud scanning on the steel structure component to be measured by a three-dimensional laser scanning device. The preset marking point refers to a position point set in advance by an engineer and capable of distinguishing the characteristics of the steel structure component, such as the maximum boundary points of the steel structure component. For example, a coordinate system is constructed with the center of the steel structure component, and multiple points such as the vertex, bottom point, left point, right point, upper left point, upper right point, etc. of the steel structure component that can express the characteristics of the steel structure component can be obtained.

[0046] Step S2: derive the point cloud coordinate value of each preset marked point based on the multiple preset marked points and point cloud data on the steel structure component to be measured, and calculate the actual distance between any two preset marked points based on the point cloud coordinate values ​​of the multiple preset marked points.

[0047] Step S3: Determine whether the steel structure component to be measured has a dimensional error based on the actual distance between any two preset marking points and the design parameters of the steel structure component to be measured. If so, execute step S4.

[0048] Specifically, the preset marking points are points pre-set on the steel structure component to be measured, and the standard distance between any two preset marking points can be derived based on the design parameters of the steel structure component to be measured. In this way, the calculated actual distance can be compared with the standard distance to determine whether there is a dimensional error in the steel structure component to be measured. Normally, within the allowable error range, the error between the actual distance between any two preset marking points of the steel structure component to be measured and the corresponding standard distance should be within a certain error range. Therefore, if the actual distance between any two preset marking points is inconsistent with the corresponding standard distance, it can be considered that the steel structure component to be measured has a dimensional error. In addition, the standard distance for the steel structure component to be measured can be calculated based on the design parameters of the steel structure component to be measured.

[0049] Step S4: construct a point cloud model of the steel structure component to be measured based on the point cloud data of the steel structure component to be measured, and match the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured to obtain a difference model, and then obtain deviation data of the point cloud model of the steel structure component to be measured relative to the design model of the steel structure component to be measured based on the difference model.

[0050] Generally speaking, when the size of the steel structure component to be measured meets the requirements, the design model of the steel structure component to be measured is a CAD model, and after a certain scaling factor, the CAD model and the point cloud model should be the same after multiple rotations and / or translations. On the contrary, when the size of the steel structure component to be measured does not meet the requirements, there will be a certain error between the point cloud model and the design model after scaling, rotation or translation. Therefore, this embodiment matches the scanned point cloud model with the design model to obtain a matched difference model. In this way, the actual deviation data of the steel structure component to be measured can be more intuitively viewed based on the difference model, that is, the deviation data of the point cloud model relative to the design model, thereby realizing the dimensional error detection of the steel structure component to be measured. This process does not require manual measurement, recording or operation, which frees up manpower and improves the efficiency of dimensional error detection. Of course, since no manual measurement or recording is required, the accuracy of dimensional error can also be improved.

[0051] More preferably, in the process of matching the point cloud model with the design model, that is, when the point cloud model and / or the design model are rotated, translated and scaled, the point cloud model of the steel structure component to be measured and the design model are converted to the same coordinate system based on the spatial matching algorithm and the coordinate conversion algorithm to achieve matching between the two. Figure 3As shown in the figure, the solid line border is the original design size of a steel structure component to be measured, and a design model is constructed; the discrete points are the original point cloud data obtained after scanning. The original point cloud data is modeled to obtain a point cloud model, and then the point cloud model is matched with the design model to obtain the difference between the actual construction size of the steel structure component to be measured and the original design size, that is, the construction size error of the steel structure component to be measured.

[0052] More preferably, step S4 in obtaining the difference model specifically includes: obtaining the difference model by rotating, translating, and scaling the point cloud model of the steel structure component to be measured multiple times so that the sum of the squares of the distances between the points in the point cloud model of the steel structure component to be measured and the points in the design model of the steel structure component to be measured is minimized.

[0053] Furthermore, when obtaining the difference model, the point cloud model and the design model are first converted to the same coordinate system, and then the following steps are performed:

[0054] Step S41: according to each preset marked point, a first coordinate value of each preset marked point is found from the point cloud model of the steel structure component to be measured.

[0055] Step S42: according to each preset marked point, find out the second coordinate value of each preset marked point from the design model of the steel structure component to be measured.

[0056] Step S43: Match the point cloud model and the design model after multiple scaling, rotation, and translation to ensure that the sum of the squares of the distances between the multiple first coordinate values ​​and the corresponding second coordinate values ​​is minimized; and when the sum of the squares of the distances between the multiple first coordinate values ​​and the corresponding second coordinate values ​​is minimized, a difference model is obtained based on the matching results of the point cloud model and the design model.

[0057] Specifically, when the actual distance measured from the pre-defined marked points does not conform to the standard dimension data, the present invention further uses model matching to detect the differences between the point cloud model and the design model, thereby deriving the deviation direction and deviation data based on the difference model. Compared to direct model matching, this approach can first remove the steel structure components to be tested that meet the requirements, and then further detect the deviation of the steel structure components that do not meet the requirements, thereby improving detection efficiency.

[0058] When obtaining the difference model, the present invention realizes model matching by selecting a plurality of preset marking points on the steel structure component to be measured, thereby avoiding low efficiency of data processing caused by an excessive number of points.

[0059] Furthermore, the deviation data includes a deviation direction and a deviation amount. The deviation direction refers to the deviation direction of the point cloud model relative to the design model, and the deviation amount refers to the deviation amount of the point cloud model relative to the design model.

[0060] Step S5: Display the difference model and deviation data to the staff through a visual display platform.

[0061] Specifically, the deviation direction, deviation amount and difference model in the deviation data are displayed to the staff through a visual display platform. In addition, when the deviation data and difference model are displayed to the staff through a visual display platform, the difference between the two is highlighted with different colors, and the deviation direction is marked in the difference model and then displayed through the visual display platform. The deviation data is made more intuitive by highlighting and marking. For example, the area with a deviation of 1 mm to 5 mm is marked in yellow, and the area with a deviation of more than 5 mm is marked in red, providing a more intuitive visual display. In this way, the staff can observe the deviation direction and deviation amount more intuitively, realize the detection of dimensional differences of the steel structure components to be measured, and at the same time, realize production adjustment and quality inspection of the steel structure components to be measured based on the deviation direction and deviation amount. This process does not require human participation, improves the detection efficiency and accuracy, and ensures the safety of the measurement personnel.

[0062] More preferably, step S5 further comprises: obtaining a corresponding adjustment scheme from a scheme matching model constructed in the system according to the type, deviation direction and deviation amount of the steel structure component to be tested, and pushing the corresponding adjustment scheme to the staff, and displaying the deviation direction, deviation amount and difference model to the staff through a visual display platform. In this case, the deviation direction, deviation amount and corresponding adjustment scheme of the same type of steel structure component to be tested are statistically analyzed, and the matching relationship between the deviation direction, deviation amount and adjustment scheme of the steel structure component to be tested is obtained by combining the machine learning model, and then the scheme matching model of the corresponding type of steel structure component is constructed based on the matching relationship. In this case, the deviation direction, deviation amount and adjustment scheme of the historical steel structure component are obtained from the historical adjustment scheme during the inspection process of the steel structure component, or are manually summarized based on work experience. The AI ​​big model and the machine learning model are used to realize data mining, so as to obtain the matching relationship between the deviation direction, deviation amount and adjustment scheme, so as to obtain the deviation direction, deviation amount and adjustment scheme of the steel structure component to be tested and push it to the staff, so as to provide the staff with relevant adjustment suggestions for reference. At the same time, when the staff has completed the adjustment of the steel structure components, the dimensional error detection method provided by the present invention can be restarted to scan the steel structure components again to re-collect and compare the data to achieve real-time analysis of the errors until all deviation values ​​meet the standards. At the same time, the adjusted model will be displayed through a visual display platform to intuitively inform the staff whether the adjustment is correct.

[0063] The dimensional error detection provided by the present invention features a high degree of automation. Operators only need to follow the system's prompts to perform necessary operations, such as scanning steel components. This significantly reduces human intervention, improves the efficiency of detection results, and reduces detection errors. Through automated error detection, real-time feedback, and calibration suggestion generation, the system helps workers quickly and accurately complete dimensional calibration, reducing human error and improving construction quality. Compared to traditional manual measurement, the system significantly reduces measurement errors and can automatically identify and mark error locations, ensuring the dimensional accuracy of installed steel components. Through automated error detection and real-time feedback, the system can complete dimensional calibration of large-scale structures in a short period of time, reducing the need for multiple measurements and verifications and improving overall construction efficiency. Furthermore, the present invention automatically generates calibration suggestions by sending adjustment plans to workers, helping them make adjustments quickly. This avoids the complexities of repeated measurements and manual calibration, significantly shortening the construction cycle. Traditional steel component calibration processes often involve high rework and adjustment costs, but this system improves construction accuracy and reduces construction costs. Furthermore, the system reduces the reliance on highly skilled surveyors, making construction operations more convenient and reducing labor costs.

[0064] The dimensional error detection system provided by the present invention is particularly suitable for various types of steel structure engineering projects, and is particularly effective during the production and installation of large, complex steel structures. For example, in high-rise buildings, dimensional calibration of steel components is particularly important, as errors accumulate at each floor, affecting overall construction accuracy and safety. This system can effectively control the installation accuracy of steel components in high-rise buildings. Through automatic error detection and real-time feedback, it ensures that the dimensions of each steel component meet design requirements, reducing overall structural deviations caused by error accumulation. For another example, in the construction of long-span bridges, the installation of steel components requires high precision and consistency. Traditional manual measurement methods struggle to meet design requirements, but this system can perform high-precision calibration of each key connection point in the bridge structure. By comparing point cloud data with the design model, it identifies structural errors and provides calibration recommendations to ensure the stability of the bridge structure. Furthermore, in buildings such as factories and warehouses, which often utilize large-scale steel components, the installation process often requires calibration of multiple connection points in a short period of time. The system's automated calibration function reduces manual measurement time, quickly generates error analysis and adjustment recommendations, and improves construction efficiency and shortens project cycles.

[0065] This invention significantly improves the accuracy and efficiency of dimensional verification of steel structural components, reduces the complexity of manual operations, and enhances construction quality and safety. Furthermore, it incorporates calibration transactions during the automatic error detection process to reduce errors caused by manual operation. A real-time feedback mechanism ensures that the adjustment accuracy meets requirements, thereby reducing rework rates and improving project quality. This invention reduces labor costs, shortens construction cycles, and provides reliable technical support for the inspection and subsequent installation of steel structural components.

[0066] Example 2

[0067] Based on the first embodiment, the present invention further provides another embodiment, a device for detecting dimensional errors of steel structure components based on point clouds, specifically comprising:

[0068] A point cloud acquisition module is used to perform point cloud scanning on the steel structure component to be measured to obtain point cloud data of the steel structure component to be measured;

[0069] A distance calculation module is used to obtain the point cloud coordinate value of each preset marked point based on multiple preset marked points and point cloud data on the steel structure component to be measured, and to calculate the actual distance between any two preset marked points based on the point cloud coordinate values ​​of the multiple preset marked points;

[0070] a judgment module, configured to judge whether the steel structure component to be measured has a dimensional error based on the actual distance between any two preset marked points and the design parameters of the steel structure component to be measured, and if so, perform a matching step;

[0071] a matching module, configured to construct a point cloud model of the steel structure component to be measured based on the point cloud data of the steel structure component to be measured, and to match the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured to obtain a difference model, and further to obtain deviation data of the point cloud model of the steel structure component to be measured relative to the design model of the steel structure component to be measured based on the difference model;

[0072] The display push module is used to display the difference model and deviation data to the staff through the visual display platform.

[0073] Example 3

[0074] Based on the first embodiment, the present invention further provides another embodiment, a computer-readable storage medium having a steel structure member dimensional error detection program stored thereon, wherein the steel structure member dimensional error detection program is a computer program, and when the steel structure member dimensional error detection program is executed by a processor, the following steps are implemented:

[0075] Point cloud acquisition steps: performing point cloud scanning on the steel structure component to be measured to obtain point cloud data of the steel structure component to be measured;

[0076] Distance calculation step: deriving the point cloud coordinate value of each preset marked point based on multiple preset marked points and point cloud data on the steel structure component to be measured, and calculating the actual distance between any two preset marked points based on the point cloud coordinate values ​​of the multiple preset marked points;

[0077] Determination step: determining whether the steel structure component to be measured has a dimensional error based on the actual distance between any two preset marked points and the design parameters of the steel structure component to be measured, and if so, executing the matching step;

[0078] Matching step: constructing a point cloud model of the steel structure component to be measured based on the point cloud data of the steel structure component to be measured, and matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured to obtain a difference model, and then obtaining deviation data of the point cloud model of the steel structure component to be measured relative to the design model of the steel structure component to be measured based on the difference model;

[0079] Display push step: Display the difference model and deviation data to the staff through the visual display platform.

[0080] Furthermore, the point cloud acquisition step specifically includes: first, marking multiple preset marking points on the steel structure component to be measured according to the characteristics of the steel structure component to be measured, and then performing point cloud scanning on the steel structure component to be measured through a three-dimensional laser scanning device to obtain point cloud data of the structural component to be measured.

[0081] Furthermore, the judgment step specifically includes: first obtaining the design parameters of the steel structure component to be measured and then deriving the standard distance between any two preset marked points of the steel structure component to be measured based on the design parameters of the steel structure component to be measured; then matching the actual distance between each two preset marked points with the corresponding standard distance, and then judging whether there is a dimensional error in the steel structure component to be measured based on the matching result; wherein, when the actual distance between any two preset marked points in the matching result does not match the corresponding standard distance, there is a dimensional error in the steel structure component to be measured.

[0082] Furthermore, matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured in the matching step specifically includes: matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured based on a spatial matching algorithm and a coordinate transformation algorithm.

[0083] Furthermore, in the matching step, matching the point cloud model of the steel structure component to be measured with the design drawing of the steel structure component to be measured to obtain a difference model specifically includes: obtaining the difference model when the sum of the squares of the distances between the points in the point cloud model of the steel structure component to be measured and the points in the design model of the steel structure component to be measured is minimized after the point cloud model of the steel structure component to be measured is scaled, translated, and rotated multiple times.

[0084] Furthermore, in the matching step, matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured to obtain a difference model further specifically includes:

[0085] Step 1: Find the first coordinate value of each preset marked point from the point cloud model of the steel structure component to be measured according to each preset marked point;

[0086] Step 2: Find the second coordinate value of each preset marked point from the design model of the steel structure component to be measured according to each preset marked point;

[0087] Step 3: Match the point cloud model and the design model after multiple scaling, rotation, and translation to ensure that the sum of the squares of the distances between the multiple first coordinate values ​​and the corresponding second coordinate values ​​is minimized; and when the sum of the squares of the distances between the multiple first coordinate values ​​and the corresponding second coordinate values ​​is minimized, a difference model is obtained based on the matching results of the point cloud model and the design model.

[0088] Furthermore, the display push step includes: obtaining the deviation direction and deviation amount of the steel structure component to be measured based on the deviation data, and displaying the difference model, deviation data and deviation amount to the staff through a visual display platform.

[0089] Furthermore, the display push step also includes: matching the corresponding adjustment scheme from the scheme matching model constructed in the system according to the type, deviation direction and deviation amount of the steel structure component to be tested, and pushing the corresponding adjustment scheme to the staff; through statistical analysis of the deviation direction, deviation amount and corresponding adjustment scheme of the same type of steel structure components in history, and combining the machine learning model to obtain the matching relationship between the deviation direction, deviation amount and adjustment scheme of the steel structure component to construct a scheme matching model for the corresponding type of steel structure component.

[0090] The above embodiments are only preferred embodiments of the present invention and cannot be used to limit the scope of protection of the present invention. Any non-substantial changes and replacements made by technicians in this field on the basis of the present invention fall within the scope of protection required by the present invention.

Claims

1. A method for detecting dimensional errors of steel structure components based on point cloud, characterized in that: The steel structure member dimensional error detection includes: Point cloud acquisition step: performing point cloud scanning on the steel structure component to be measured to obtain point cloud data of the steel structure component to be measured; Distance calculation step: obtaining the point cloud coordinate value of each preset marked point according to the plurality of preset marked points and the point cloud data on the steel structure member to be measured, and calculating the actual distance between any two preset marked points according to the point cloud coordinate values ​​of the plurality of preset marked points; Determining step: determining whether the steel structure component to be measured has a dimensional error based on the actual distance between any two preset marked points and the design parameters of the steel structure component to be measured, and if so, executing the matching step; the determining step specifically comprises: first obtaining the design parameters of the steel structure component to be measured and then deriving the standard distance between any two preset marked points of the steel structure component to be measured based on the design parameters of the steel structure component to be measured; then matching the actual distance between every two preset marked points with the corresponding standard distance, and then determining whether the steel structure component to be measured has a dimensional error based on the matching result; wherein, when the actual distance between any two preset marked points in the matching result does not match the corresponding standard distance, the steel structure component to be measured has a dimensional error; Matching step: constructing a point cloud model of the steel structure component to be measured based on the point cloud data of the steel structure component to be measured, and matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured to obtain a difference model, and then obtaining deviation data of the point cloud model of the steel structure component to be measured relative to the design model of the steel structure component to be measured based on the difference model; The matching step of matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured specifically includes: matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured based on a spatial matching algorithm and a coordinate transformation algorithm; the matching step of obtaining a difference model based on the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured specifically includes: obtaining a difference model when the sum of the squares of the distances between the points in the point cloud model of the steel structure component to be measured and the points in the design model of the steel structure component to be measured is minimized after scaling, translating, and rotating the point cloud model of the steel structure component to be measured multiple times; The matching step of matching the point cloud model of the steel structure component to be measured with the design model of the steel structure component to be measured to obtain a difference model further specifically includes: Step 1: according to each preset marked point, find out the first coordinate value of each preset marked point from the point cloud model of the steel structure component to be measured; Step 2: according to each preset marked point, find out the second coordinate value of each preset marked point from the design model of the steel structure component to be measured; Step 3: Matching the point cloud model and the design model after multiple scaling, rotation, and translation to ensure that the sum of the squares of the distances between the multiple first coordinate values ​​and the corresponding second coordinate values ​​is minimized; and obtaining a difference model based on the matching results of the point cloud model and the design model when the sum of the squares of the distances between the multiple first coordinate values ​​and the corresponding second coordinate values ​​is minimized; Display push step: displaying the difference model and deviation data to staff through a visual display platform.

2. The method for detecting dimensional errors of steel structure components based on point cloud according to claim 1, characterized in that: The point cloud acquisition step specifically includes: first, marking a plurality of preset marking points on the steel structure component to be measured according to the characteristics of the steel structure component to be measured, and then performing point cloud scanning on the steel structure component to be measured by a three-dimensional laser scanning device to obtain point cloud data of the steel structure component to be measured.

3. The method for detecting dimensional errors of steel structure components based on point cloud according to claim 1, characterized in that: The display push step includes: obtaining the deviation direction and deviation amount of the steel structure component to be measured according to the deviation data, and displaying the difference model, deviation data and deviation amount to the staff through a visual display platform.

4. The method for detecting dimensional errors of steel structure components based on point cloud according to claim 3, characterized in that: The display push step also includes: matching the corresponding adjustment scheme from the scheme matching model constructed in the system according to the type, deviation direction and deviation amount of the steel structure component to be tested, and pushing the corresponding adjustment scheme to the staff; through statistical history of the deviation direction, deviation amount and corresponding adjustment scheme of the same type of steel structure components, and combining the machine learning model to mine the matching relationship between the deviation direction, deviation amount and adjustment scheme of the steel structure component to construct a scheme matching model for the corresponding type of steel structure component.

5. A device for detecting dimensional errors of steel structure components based on point cloud, applied to the method for detecting dimensional errors of steel structure components based on point cloud according to any one of claims 1 to 4, characterized in that: include: A point cloud acquisition module is used to perform point cloud scanning on the steel structure component to be measured to obtain point cloud data of the steel structure component to be measured; a distance calculation module, configured to obtain a point cloud coordinate value of each preset marked point based on the plurality of preset marked points and the point cloud data on the steel structure member to be measured, and to calculate the actual distance between any two preset marked points based on the point cloud coordinate values ​​of the plurality of preset marked points; a judgment module, configured to judge whether the steel structure component to be measured has a dimensional error based on the actual distance between any two preset marked points and the design parameters of the steel structure component to be measured, and if so, execute a matching module; a matching module, configured to construct a point cloud model of the steel structure component to be measured based on the point cloud data of the steel structure component to be measured, and match the point cloud model of the steel structure component to be measured with a design model of the steel structure component to be measured to obtain a difference model, and further obtain deviation data of the point cloud model of the steel structure component to be measured relative to the design model of the steel structure component to be measured based on the difference model; The display push module is used to display the difference model and deviation data to the staff through a visual display platform.

6. A computer-readable storage medium storing a steel structure member dimensional error detection program, characterized in that: The steel structure component dimensional error detection program is a computer program. When the steel structure component dimensional error detection program is executed by a processor, the steps of the steel structure component dimensional error detection method based on point cloud as described in any one of claims 1 to 4 are implemented.

Images