Method for determining center coordinates of lug plate through holes of lug plate group in steel structure installation engineering based on three-dimensional point cloud

Abstract

The invention discloses a three-dimensional point cloud-based ear plate through hole center coordinate determination method for a steel structure installation engineering ear plate group. The method comprises the following steps of: acquiring overall three-dimensional point cloud data of the steel structure installation engineering ear plate group, and performing spatial transformation, denoising and filtering from an acquisition coordinate system to a construction site coordinate system on the overall three-dimensional point cloud data; the method comprises the steps of lug plate component node local point cloud segmentation, lug plate plane feature fitting, lug plate point cloud attitude adjustment, lug plate through hole contour feature extraction and lug plate through hole center coordinate calculation for the denoised and filtered overall three-dimensional point cloud data of the lug plate group in the steel structure installation project. And determining the actually measured coordinate position of the center coordinate of each lug plate through hole in the lug plate group in the steel structure installation project. According to the method, the actually-measured coordinate positions of the center coordinates of the massive lug plate through holes in the lug plate group of the steel structure installation engineering are automatically determined in batches through the three-dimensional point cloud data, and the method has the effects of high efficiency, short time consumption and high precision in determining the actually-measured coordinate positions of the center coordinates of the massive lug plate through holes in the lug plate group.

Description

Technical Field

[0001] The present invention relates to the technical field of building installation engineering, and particularly relates to a method for determining the center coordinates of the through holes of a group of gusset plates in a steel structure installation project based on three-dimensional point clouds. Background Art

[0002] In the field of steel structure installation engineering, gusset plates are key components for spatial node connection and curtain wall support systems. Their spatial positioning accuracy directly determines the construction quality of adjacent components and the safety performance of the structural system. In large public building projects, the number of gusset plate groups often reaches the scale of tens of thousands, and they have characteristics such as small through hole diameters and low installation tolerances, which affect the connection construction of fasteners such as pins or bolts. Therefore, it is urgent to establish an efficient and accurate hole position measurement technology system to determine the center coordinates of the through holes of each gusset plate in the gusset plate group for the construction quality review of gusset plates and the detailed design of secondary components. In this field, the total station is usually used to measure the center coordinates of the through holes of gusset plates point by point on site, that is, reflective patches are pasted at the through holes of each gusset plate in the steel structure project ear plate group in advance, and then the center points of the patches are aimed at one by one and their coordinates are measured. However, due to the limited field of view of the total station, it is extremely difficult for manual visual search of the target and aiming measurement, the measurement efficiency is extremely low, the single-point measurement takes up to 5 - 6 minutes, which is extremely likely to cause visual fatigue of the operators and the accumulation of visual aiming errors, and the accuracy of the obtained center coordinates of the through holes of the gusset plates is low. In this field, measurement robots equipped with automatic target aiming functions are also used to determine the center coordinates of the through holes of gusset plates. Its disadvantages are: although it solves the problems of automatic recognition, precise aiming and tracking locking of measurement targets such as prisms and reflective patches, the equipment cost is relatively high, it has a good measurement effect on discrete gusset plates, but has a poor measurement effect, low accuracy and large error for gusset plate groups in the target array scenarios of high-density steel structure installation projects such as curtain wall support systems, and still a large amount of time is required to paste reflective patches at the through holes of each gusset plate in the steel structure project gusset plate group. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for determining the center coordinates of the through holes of a group of gusset plates in a steel structure installation project based on three-dimensional point clouds, so as to solve the problems of low efficiency, low accuracy and long time consumption in determining the center coordinates of the through holes of each gusset plate in the steel structure project gusset plate group by traditional methods.

[0004] To solve the above technical problems, the present invention provides a method for determining the center coordinates of the through holes of a group of gusset plates in a steel structure installation project based on three-dimensional point clouds, including:

[0005] Obtaining the overall three-dimensional point cloud data of the gusset plate group in the steel structure installation project;

[0006] Perform a spatial transformation on the coordinate system for obtaining the overall three-dimensional point cloud data of the ear plate group in the steel structure installation project to the construction site coordinate system, and perform denoising and filtering on the overall three-dimensional point cloud data of the ear plate group in the steel structure installation project after the spatial transformation;

[0007] For the overall three-dimensional point cloud data of the ear plate group in the steel structure installation project after denoising and filtering, through the steps of local point cloud segmentation of the ear plate component nodes, fitting of the ear plate plane features, adjustment of the ear plate point cloud attitude, extraction of the ear plate through-hole contour features, and calculation of the center coordinates of the ear plate through-holes, determine the measured coordinate positions of the center coordinates of each ear plate through-hole in the ear plate group of the steel structure installation project.

[0008] Furthermore, the method for determining the center coordinates of the ear plate through-holes in the ear plate group of the steel structure installation project based on three-dimensional point cloud provided by the present invention further includes:

[0009] Check whether the measured coordinates of the center coordinates of each ear plate through-hole are valid by comparing the differences in the measured coordinates of the center coordinates of the through-holes on both sides in the thickness direction of each ear plate in the ear plate group of the steel structure installation project after the spatial transformation;

[0010] Calculate the deviations between the measured coordinates of the center coordinates of the through-holes on both sides in the thickness direction of each ear plate in the effective ear plate group of the steel structure installation project after the spatial transformation and the design coordinates respectively, and take the average value as the installation deviation of each ear plate for coloring and three-dimensional visualization display.

[0011] Furthermore, the method for determining the center coordinates of the ear plate through-holes in the ear plate group of the steel structure installation project based on three-dimensional point cloud provided by the present invention colors the point cloud data of the measured coordinates of the center coordinates of the ear plate through-holes in a single color or in color from light to deep or from deep to light according to the degree of deviation from small to large or from large to small, so as to distinguish the degree of deviation of the measured coordinates of the center coordinates of each ear plate through-hole through a chromatogram.

[0012] Furthermore, the method for determining the center coordinates of the ear plate through-holes in the ear plate group of the steel structure installation project based on three-dimensional point cloud provided by the present invention, the method for local point cloud segmentation of the ear plate component nodes includes:

[0013] Step 311: Obtain the BIM design model of the target building and the ear plate group in the steel structure installation project thereon, and batch extract and export the set of center point pairs of the designed coordinates of the ear plate through-holes in the ear plate group of the steel structure installation project on the BIM design model;

[0014] M = {(M 1L , M 1R ), (M 2L , M 2R ), …, (M iL , M iR ), …, (M nL , M nR)}

[0015] Among them, the center point pair (M iL , M iR ) represents the center points on the left and right side surfaces of the through holes of the i-th ear plate;

[0016] Step 312: Load the set of center point pairs of the ear plate through hole design coordinates of the ear plate group of the steel structure installation project exported from the BIM design model onto the overall three-dimensional point cloud data PC of the ear plate group of the steel structure installation project after denoising and filtering. Taking any point of the center point pair (M iL , M iR ) as the center, search and extract the neighboring points within the radius R in the overall three-dimensional point cloud data PC of the ear plate group of the steel structure installation project after denoising and filtering to obtain the local point cloud PC i of the ear plate component node M i , realizing the local point cloud segmentation of each ear plate component node in the overall three-dimensional point cloud data of the ear plate group of the steel structure installation project after denoising and filtering; where the search radius R = max(0.1, D), and D is the maximum size of the ear plate component node.

[0017] Furthermore, for the method for determining the center coordinates of the ear plate through holes of the ear plate group of the steel structure installation project based on three-dimensional point cloud provided by the present invention, the method for fitting the ear plate plane features includes:

[0018] Step 321: Fit the maximum plane feature in the local point cloud PC i of each ear plate component node after the local point cloud segmentation of the ear plate component node based on the improved random sample consensus algorithm; specifically including: i Step 321.1: Set the initial inlier distance threshold [d] and the maximum number of iterations N, and initialize the best fitting plane;

[0019] Step 321.2: Randomly extract three non-collinear points from the local point cloud PC

[0020] of the ear plate component node to generate a candidate plane, calculate the Euclidean distance from all data points in the local point cloud to the candidate plane, and select the points with Euclidean distance less than the distance threshold [d] as inliers; i Step 321.3: Count the number of inliers of the candidate plane described in Step 321.2. If it is greater than the number of inliers of the best fitting plane, update the candidate plane as the best fitting plane, otherwise do not update;

[0021] Step 321.4: Calculate the number of inliers of the current best fitting plane in the local point cloud PC

[0022] Step 321.4: Calculate the number of inliers of the current best fitting plane in the local point cloud PC iThe proportion p in it, and dynamically adjust the inlier distance threshold according to the iterative formula (1).

[0023] d = d×(1 - α×p) (1);

[0024] In formula (1), α is the threshold adjustment coefficient, and 0 < α < 1;

[0025] Step 321.5: Repeat steps 321.2 to 2.1.4 until the maximum number of iterations N described in step 321.1 is reached or the best - fitting plane has not been updated for 10 consecutive iterations, to determine the best - fitting plane as the local point cloud PC i of the largest plane, and output its plane characteristic parameters (A i , B i , C i , D i ) and the inlier set pc i .

[0026] Furthermore, for the method for determining the center coordinates of the ear - plate through - holes of the ear - plate group in the steel - structure installation project based on three - dimensional point cloud provided by the present invention, the method for adjusting the attitude of the ear - plate point cloud includes:

[0027] Step 331: Traverse the points P i in the inlier set pc i (x i , y i , z i ) of the largest plane described in step 321.5, and project P i onto the largest plane, and the coordinates of the projection point P i ′(x i ′, y i ′, z i ′) of the largest plane obtained by formula (2) are:

[0028]

[0029] Step 332: Calculate and determine the rotation axis k and rotation angle θ for aligning the largest plane described in step 331 to the XOY plane, and construct the rotation matrix R accordingly;

[0030] Step 333: Perform a rotation transformation on the projection point set {P i ′} according to step 332 to adjust the point - cloud attitude, and obtain the point cloud pc i ′, realizing the adjustment of the ear - plate point - cloud attitude.

[0031] Furthermore, for the method for determining the center coordinates of the ear - plate through - holes of the ear - plate group in the steel - structure installation project based on three - dimensional point cloud provided by the present invention, the rotation axis k, rotation angle θ, and rotation matrix R are calculated according to formulas (3), (4), and (5) respectively:

[0032] k = (A i , B i , C i ) × (0, 0, 1) = (B i , -A i , 0) (3);

[0033]

[0034] In Equation (4), ‖(A i , B i , C i )‖ is the vector length;

[0035] In Equation (5), I is the third-order identity matrix, and [k]× is the skew-symmetric matrix of k, where the skew-symmetric matrix of k is:

[0036]

[0037] Furthermore, the method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on the three-dimensional point cloud provided by the present invention, the method for extracting the contour features of the ear plate through holes includes:

[0038] Step 341: Use the concave hull algorithm to calculate the minimum convex polygon of the point cloud pc i ′ after the attitude adjustment of the ear plate point cloud in Step 333, and gradually optimize the outer boundary convex polygon and identify the internal hole area according to the local density information of the point cloud and the adjacent relationship of the connection points, and then connect the hole edge points in sequence to output the inner and outer contour point sets L all ;

[0039] Step 342: Use the alpha-shape algorithm to construct a rolling circle with a radius of alpha, and make it roll around the point cloud pc i ′ after the attitude correction in Step 333, and output the data point set formed by the points in the vicinity of the movement trajectory of the rolling circle and the point cloud pc i ′ to form the outer contour point set L outer ;

[0040] Step 343: Perform a Boolean difference set operation on the inner and outer contour point set L all in Step 341 and the outer contour point set L outer in Step 342 to solve the inner contour point set L inner of the point cloud; if there are multiple through holes in the ear plate component, set a distance threshold d0, and perform Euclidean clustering on the inner contour point set L inner of the ear plate to obtain the contour features {L j} of each ear plate through hole.

[0041] Further, the present invention provides a method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on 3D point clouds. The method for obtaining the center coordinates of the ear plate through holes includes:

[0042] Step 5: According to the contour feature L of the ear plate through hole j , solve the center coordinates of the ear plate through hole, and obtain the measured coordinates of the center of the ear plate through hole according to the inverse transformation of the projection transformation relationship described in step 333.

[0043] Further, the present invention provides a method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on 3D point clouds. The method for solving the center coordinates of the ear plate through hole includes:

[0044] If the shape of the ear plate through hole is a circle with a known diameter, then based on the least squares method, fit the inner contour point set, and solve the center coordinates of the circle as the center coordinates of the ear plate through hole;

[0045] If the shape of the ear plate through hole is a circle or an ellipse with unknown geometric parameters, then based on the RANSAC algorithm, fit the inner contour point set, and solve the center coordinates of the inner contour point set of the circle or ellipse with unknown geometric parameters after fitting as the center coordinates of the ear plate through hole;

[0046] If the shape of the ear plate through hole is an irregular shape, then directly solve the centroid coordinates of the inner contour point set of the irregular shape as the center coordinates of the ear plate through hole.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0048] The method for determining the center coordinates of the ear plate through-holes in the ear plate group of a steel structure installation project based on 3D point cloud provided by the present invention determines the measured coordinate positions of the center coordinates of the through-holes of each ear plate in the ear plate group of the steel structure installation project by obtaining the overall 3D point cloud data of the ear plate group of the steel structure installation project, performing spatial transformation, denoising, and filtering on its coordinate system, and then through steps such as local point cloud segmentation of the ear plate component nodes, fitting of the ear plate plane features, adjustment of the ear plate point cloud posture, extraction of the ear plate through-hole contour features, and calculation of the center coordinates of the ear plate through-holes. Thus, the measured coordinate positions of the center coordinates of a large number of ear plate through-holes in the ear plate group of the steel structure installation project are determined batch by batch and automatically through the 3D point cloud data, achieving the effects of high efficiency and short time consumption in determining the measured coordinates of the center coordinates of a large number of ear plate through-holes in the ear plate group; through the extraction of the ear plate through-hole features from the 3D point cloud data and the calculation of the center coordinates of the ear plate through-holes, the rapid calculation of the measured coordinates of the center coordinates of a large number of ear plate through-holes in the ear plate group is realized, achieving the effects of high efficiency and high precision. It avoids the problems of long time consumption and low efficiency caused by pasting reflection patches at the through-holes of each ear plate in the ear plate group of the steel structure project at the construction site in physical measurement methods such as total station and measurement robots equipped with automatic target aiming functions; it avoids the problems of low precision, easy fatigue, and limited field of vision caused by manual measurement with a total station, and avoids the problems of poor measurement effect, low precision, and large error in the measurement of the ear plate group in the target array scenario of a high-density steel structure installation project by a measurement robot equipped with an automatic target aiming function. Description of the Drawings

[0049] Figure 1 is a flowchart of the method for determining the center coordinates of the ear plate through-holes in the ear plate group of a steel structure installation project based on 3D point cloud according to an embodiment;

[0050] Figure 2 is a flowchart of determining the measured coordinate positions of the center coordinates of the through-holes of each ear plate in the ear plate group of the steel structure installation project through the overall 3D point cloud data of the ear plate group of the steel structure installation project after denoising and filtering;

[0051] Figure 3 is a schematic structural diagram of the connection relationship between the ear plate and its steel purlin;

[0052] Figure 4 is a 3D point cloud diagram of a certain ear plate component;

[0053] Figure 5 is a schematic diagram of the inner and outer contour features of a certain ear plate point cloud;

[0054] Figure 6 is a schematic diagram of extracting the center of a certain ear plate through-hole;

[0055] Figure 7 is a flowchart of the method for determining the center coordinates of the ear plate through-holes in the ear plate group of a steel structure installation project based on 3D point cloud according to another embodiment;

[0056] Figure 8 It is a flowchart for verifying the validity of the central coordinates of the through-holes on the ear plates.

[0057] Figure 9 It is a point cloud diagram of some ear plates with deviations in the measured coordinates of the central coordinates of the through-holes on the ear plates in a group of ear plates.

[0058] Figure 10 is Figure 9 the ear plate with deviation and its coloring diagram in

[0059] As shown in the figure:

[0060] 1. Steel purlin, 2. Ear plate, 3. Through-hole. Specific implementation manner

[0061] The present invention will be described in detail below with reference to the accompanying drawings: According to the following description, the advantages and features of the present invention will be clearer. It should be noted that the accompanying drawings are all in a very simplified form and use non-precise scales, only for the purpose of conveniently and clearly assisting in explaining the embodiments of the present invention.

[0062] Please refer to Figures 1 to 6 , an embodiment of the present invention provides a method for determining the central coordinates of through-holes on ear plates in a group of ear plates for a steel structure installation project based on three-dimensional point clouds, including:

[0063] Step 100, obtaining the overall three-dimensional point cloud data of the ear plate group for the steel structure installation project. The overall three-dimensional point cloud data of the ear plate group is dense point cloud data, and the overall three-dimensional point cloud data of the ear plate group on the steel structure installation project on the target building can be obtained through three-dimensional laser scanning or unmanned aerial vehicle oblique photography. For example: adopting an integrated intelligent measuring station layout algorithm and a station-mounted three-dimensional laser scanning device, first, through an automatic station layout algorithm based on the BIM model, dynamically place measuring stations in the BIM model of the object to be measured, set scanning parameters to simulate and analyze the effective scanning range of the measuring stations, so as to determine the optimal measuring station layout scheme for full coverage and high-density scanning of complex steel structures; then set up scanning targets at the measurement site, and through a three-dimensional scanning device equipped with a high-precision laser sensor, capture and output the three-dimensional point cloud data of the object to be measured. Compared with obtaining three-dimensional point cloud data by single-point measurement with a total station, it has the advantages of fast acquisition speed, high efficiency, and global acquisition.

[0064] Step 200, performing a spatial transformation on the obtained coordinate system of the overall three-dimensional point cloud data of the ear plate group for the steel structure installation project to the construction site coordinate system, and denoising and filtering the overall three-dimensional point cloud data of the ear plate group for the steel structure installation project after the spatial transformation. The obtained coordinate system includes the scanning coordinate system of the three-dimensional laser scanning device and the photography coordinate system of the unmanned aerial vehicle oblique photography in the above two acquisition methods.

[0065] Among them, spatial transformation, denoising, and filtering are preprocessing operations on the overall three-dimensional point cloud data of the ear plate group, which include intelligent registration, coordinate system conversion, and noise filtering functions, and are used to preprocess the overall three-dimensional point cloud data of the ear plate group. Specifically, it can include: First, through the sampling consistency initial registration algorithm and the iterative closest point algorithm, accurate registration between multi-station point clouds is achieved to obtain the overall point cloud of the object to be measured; Secondly, through the target recognition and fitting algorithm, measurement targets such as target balls or planar targets in the overall point cloud are automatically recognized, and the center coordinates of the reference control points are obtained by fitting the target center based on features such as shape and color. Then, based on the real-world coordinates / construction site coordinates of the reference control points, the three-dimensional point cloud is spatially transformed from the scanning coordinate system to the real-world coordinate system / construction site coordinate system; Finally, based on the target point cloud extraction algorithm of the BIM model of the object to be measured, the BIM model of the target object is downsampled into point clouds, and the target object point cloud is obtained by searching for neighboring points based on it, thereby removing the remaining non-target object points. Then, based on methods such as Euclidean clustering and statistical filtering, the outlier noise points in the object point cloud are effectively filtered, and the accurately denoised target object point cloud is output to form the overall three-dimensional point cloud data of the ear plate group of the steel structure installation project after spatial transformation, denoising, and filtering.

[0066] Step 300: Determine the measured coordinate positions of the center coordinates of each ear plate through-hole in the ear plate group of the steel structure installation project from the overall three-dimensional point cloud data of the ear plate group of the steel structure installation project after denoising and filtering through the steps from Step 310 to Step 350. Among them:

[0067] Step 310: Local point cloud segmentation of the ear plate component nodes, as Figure 4 shown. Among them Figure 4 illustrates the three-dimensional point cloud diagram of a certain ear plate component. Specifically, it includes:

[0068] Step 311: Obtain the BIM design model of the target building and the ear plate group of the steel structure installation project on it, and batch extract and export the set of center point pairs of the designed coordinates of the ear plate through-holes in the ear plate group of the steel structure installation project on the BIM design model:

[0069] M = {(M 1L , M 1R ), (M 2L , M 2R ), …, (M iL , M iR ), …, (M nL , M nR )}.

[0070] Among them, the center point pair (M iL , M iR ) of the designed coordinates represents the center points on the left and right sides of the i-th ear plate through-hole.

[0071] Step 312: Load the center point pair set of the ear plate through hole design coordinates of the steel structure installation engineering ear plate group derived from the BIM design model into the de-noised and filtered steel structure installation engineering ear plate group overall three-dimensional point cloud data PC, and use the center point pair (M iL ,M iR ) as the center, the overall three-dimensional point cloud data PC of the ear plate group of the steel structure installation engineering after denoising and filtering is searched and the neighboring points within the radius R are extracted to obtain the ear plate component node M i Local point cloud PC i , realizing the local point cloud segmentation of each ear plate component node in the overall three-dimensional point cloud data of the ear plate group of the steel structure installation project after denoising and filtering; wherein the search radius R=max(0.1,D), wherein D is the maximum size of the ear plate component node.

[0072] Step 320: Fitting the ear plate plane features. Specifically, it may include:

[0073] Step 321: Fitting the local point cloud of the ear plate component nodes after segmentation based on the improved random sampling consensus algorithm (RANSAC) i Local point cloud PC i The largest planar feature in ; specifically including:

[0074] Step 321.1: Set the initial intra-point distance threshold [d] and the maximum number of iterations N, and initialize the best fitting plane. The best fitting plane refers to the plane with the largest number of intra-points in the current game.

[0075] Step 321.2: PC from the local point cloud of the ear plate component node i Three non-collinear points are randomly selected to generate a candidate plane, the Euclidean distance from all data points in the local point cloud to the candidate plane is calculated, and the points whose Euclidean distance is less than the distance threshold [d] are selected as in-place points.

[0076] Step 321.3: Count the number of in-place points of the candidate plane in step 321.2. If it is greater than the number of in-place points of the best-fit plane, update the candidate plane to the best-fit plane; otherwise, do not update it.

[0077] Step 321.4: Calculate the number of points in the local point cloud PC of the current best fitting plane i The proportion p in the local point is calculated, and the intra-point distance threshold is dynamically adjusted according to the iterative formula (1).

[0078] d = d × (1 - α × p) (1).

[0079] In formula (1), α is the threshold adjustment coefficient, 0<α<1.

[0080] Step 321.5: Repeat steps 321.2 to 2.1.4 until the maximum number of iterations N described in step 321.1 is reached or the best fit plane is not updated for 10 consecutive iterations, so as to determine that the best fit plane is the local point cloud PC. i The maximum plane of the output plane feature parameters (A i ,B i ,C i ,D i ) and the intra-station point set pc i .

[0081] The plane characteristic parameter (A i ,B i ,C i ,D i ) is the plane after the ear plate plane features are fitted.

[0082] Step 330: Adjust the ear plate point cloud posture. Specifically, it may include:

[0083] Step 331: Traverse the local point set pc of the maximum plane described in step 321.5 i Point P in i (x i ,y i ,z i ), and P i Projecting onto the maximum plane, we get the projection point P of the maximum plane of formula (2) i ′(x i ′,y i ′,z i ′) is:

[0084]

[0085] Where P i ′ describes the point coordinate transformation method, that is, how to project the local points of the plane onto the largest plane, and transform an approximate plane point cloud into a strict plane according to the projection point coordinates, which is used in the subsequent steps to extract boundary features from the two-dimensional point cloud.

[0086] Step 332: Calculate and determine the rotation axis k and rotation angle θ of the maximum plane aligned to the XOY plane in step 331, and construct a rotation matrix R accordingly; wherein the rotation axis k is calculated according to formula (3):

[0087] k=(A i ,B i ,C i )×(0,0,1)=(B i ,-A i ,0) (3).

[0088] Wherein the rotation angle θ is calculated according to formula (4):

[0089]

[0090] In formula (4), ‖(A i , B i , C i )‖ is the vector length.

[0091] Wherein the rotation matrix R is calculated according to formula (5):

[0092]

[0093] In formula (5), I is a third-order identity matrix, and [k] × is the skew-symmetric matrix of k, where the skew-symmetric matrix of k is:

[0094]

[0095] Step 333: Rotate the set of projection points {P i ′} according to Step 332 to adjust the point cloud attitude, and obtain the point cloud pc i ′, realizing the adjustment of the ear plate point cloud attitude.

[0096] Step 340: Extract the contour features of the ear plate through holes, as Figure 5 shown. Among them Figure 5 illustrates the inner and outer contour features of the point cloud of an ear plate. Specifically, it includes:

[0097] Step 341: Use the ConcaveHull algorithm to calculate the minimum convex polygon of the point cloud pc i ′ after the attitude adjustment of the ear plate point cloud in Step 333, and gradually optimize the outer boundary convex polygon and identify the internal hole area according to the local density information of the point cloud and the adjacent relationship of the connecting points, and then connect the hole edge points in sequence to output the inner and outer contour point set L all .

[0098] Step 342: Use the alpha-shape algorithm to construct a rolling circle with a radius of alpha, and make it roll around the point cloud pc i ′ after the attitude correction in Step 333, and output the data point set formed by the points in the vicinity of the movement trajectory of the rolling circle and the point cloud pc i ′ as the outer contour point set L outerAmong them, the radius alpha of the rolling circle can take values from 1.5 to 2 times the average density of the point cloud. The Alpha-shape algorithm is a classic method for generating the concave hull (concave shell) of a point set based on geometric parameters (α value). By controlling the radius of the disk, it captures the topological structure and concave features of the point set. Its core idea is: "rolling" a disk with a radius of α outside the point set and retaining all the points that can be contacted by the disk as the concave hull boundary.

[0099] Step 343: For the inner and outer contour point sets L described in step 341 all and the outer contour point set L described in step 342 outer perform a Boolean difference operation to solve for the inner contour point set L of the point cloud inner ; if there are multiple through-holes in the ear plate component, set a distance threshold d0, and perform Euclidean clustering on the inner contour point set L of the ear plate inner to obtain the contour features {L j} of each ear plate through-hole.

[0100] Step 350: Calculate the center coordinates of the ear plate through-holes, as Figure 6 shown. Among them Figure 6 an example of a two-dimensional point cloud of an ear plate and the center of its through-hole is illustrated. Specifically, it includes:

[0101] Step 351: According to the contour features L of the ear plate through-hole j , calculate the center coordinates of the ear plate through-hole, and based on the inverse transformation of the projection transformation relationship described in step 333, obtain the measured coordinates of the center of the ear plate through-hole. The method for calculating the center coordinates of the ear plate through-hole may include:

[0102] If the shape of the ear plate through-hole is a circle with a known diameter, then fit the inner contour point set based on the least squares method, and solve for the center coordinates of the circle as the center coordinates of the ear plate through-hole.

[0103] If the shape of the ear plate through-hole is a circle or an ellipse with unknown geometric parameters, then fit the inner contour point set based on the RANSAC algorithm, and solve for the center coordinates of the inner contour point set after fitting the circle or ellipse with unknown geometric parameters as the center coordinates of the ear plate through-hole.

[0104] If the shape of the ear plate through-hole is an irregular shape, then directly solve for the centroid coordinates of the inner contour point set of the irregular shape as the center coordinates of the ear plate through-hole. Among them, the centroid coordinates are an approximate solution of the center coordinates of the ear plate through-hole.

[0105] The method for determining the center coordinates of the earplate through-holes of the earplate group in the steel structure installation project based on 3D point cloud provided by the embodiments of the present invention determines the measured coordinate positions of the center coordinates of the through-holes of each earplate in the earplate group of the steel structure installation project by obtaining the overall 3D point cloud data of the earplate group in the steel structure installation project, performing spatial transformation, denoising and filtering on the coordinate system, and through steps such as local point cloud segmentation of the earplate component nodes, earplate plane feature fitting, earplate point cloud attitude adjustment, extraction of the contour features of the earplate through-holes, and calculation of the center coordinates of the earplate through-holes. Thus, the measured coordinate positions of the center coordinates of a large number of earplate through-holes in the earplate group of the steel structure installation project are determined batchwise and automatically through 3D point cloud data, achieving the effects of high efficiency and short time consumption in determining the measured coordinates of the center coordinates of a large number of earplate through-holes in the earplate group; through the extraction of the through-hole features of the 3D point cloud data and the calculation of the center coordinates of the earplate through-holes, the rapid calculation of the measured coordinates of the center coordinates of a large number of earplate through-holes in the earplate group is realized, with the effects of high efficiency and high precision. It avoids the problems of time-consuming and low efficiency caused by pasting reflection patches at the through-holes of each earplate in the earplate group of the steel structure project at the construction site in physical measurement methods such as total station and measurement robots equipped with automatic target aiming functions; it avoids the problems of low precision, easy fatigue, and limited vision caused by manual measurement with a total station, and avoids the problems of poor measurement effect, low precision, and large error in measuring the earplate group in the target array scenario of a high-density steel structure installation project by a measurement robot equipped with an automatic target aiming function.

[0106] Please refer to Figures 7 to 10 and in combination with Figures 3 to 6 , in order to adjust the position of the earplates in the earplate group of the steel structure installation project at the construction site when there are deviations and ensure the connection quality at the earplates. The method for determining the center coordinates of the earplate through-holes of the earplate group in the steel structure installation project based on 3D point cloud provided by the embodiments of the present invention may further include:

[0107] Step 400, determining the installation deviation of each earplate and coloring and three-dimensionally visualizing it. Specifically, it includes:

[0108] Step 410: Checking whether the measured coordinates of the center coordinates of the through-holes of each earplate are valid by comparing the difference between the measured coordinates of the center coordinates of the through-holes on both sides of each earplate in the earplate group of the steel structure installation project after spatial transformation in the thickness direction.

[0109] In order to check whether the measured coordinates of the center coordinates of the through-holes of each earplate are valid, it may include:

[0110] Step 411: Judging whether the difference d1 between the center coordinate of the through-hole on any side of the center coordinate of the earplate through-hole and the corresponding design coordinate is less than the preset threshold Δd1, where Δd1 may be 1.1 times the maximum estimated installation deviation. If d1 < Δd1, then it is judged that this condition is valid; otherwise, it is invalid.

[0111] Step 412: Determine whether the distance d2 between the two sides of the center coordinates of the earplate through-hole is less than the preset distance threshold Δd2, where Δd2 can be 1.2 times the designed thickness of the earplate component. If d2 < Δd2, it is judged to be valid under this condition; otherwise, it is invalid.

[0112] Step 413: When d1 < Δd1 and d2 < Δd2, it is judged that the measured coordinates of the center coordinates of each earplate through-hole are valid; otherwise, they are invalid, indicating that there is an error in solving the measured coordinates of the center coordinates of the earplate through-hole, and the user is reminded to manually review.

[0113] That is to say, by comparing the difference between the center coordinates of the through-holes on both sides of each earplate in the thickness direction in the earplate group of the steel structure installation project after spatial transformation with the actual earplate thickness to verify the credibility of the coordinate validity, and by calculating the deviation between the center coordinate of any through-hole of each earplate and the design coordinate data and comparing it with the maximum estimated installation deviation to verify the credibility of the coordinate validity; when both credibilities are satisfied, it is judged to be valid.

[0114] Step 420: Calculate the deviation between the measured coordinates of the center coordinates of the through-holes on both sides of each earplate in the effective earplate group of the steel structure installation project after spatial transformation and the design coordinates, and take the average value as the installation deviation of each earplate for coloring and three-dimensional visualization display. Among them, the installation deviation data of the measured coordinates of the center coordinates of each earplate through-hole can be exported in tabular form to facilitate the deviation adjustment and correction of the installation position of the earplate at the corresponding position on the construction site. That is, according to the given construction deviation coloring scheme, based on the average deviation value between the measured value and the designed value of the center coordinates of the earplate through-hole, the three-dimensional point cloud of the affiliated earplate is colored, and a report of the earplate numbers exceeding the allowable construction deviation is generated for the user to manually review and correct on-site. Through the earplate number report, the earplate at the construction site corresponding to the corresponding deviation can be quickly found to take deviation correction measures such as adjustment. To avoid problems where fasteners such as bolts and pins cannot be connected through.

[0115] Please refer to Figure 8 , in order to identify the deviation degree of the earplate to facilitate the deviation correction of the earplate position on the construction site, the method for determining the center coordinates of the earplate through-hole in the earplate group of the steel structure installation project based on the three-dimensional point cloud provided by the embodiment of the present invention may further include:

[0116] Step 430: Color the point cloud data of the measured coordinates of the center coordinates of the earplate through-hole in the order of increasing or decreasing installation deviation in a single color or in color from light to deep or from deep to light, so as to distinguish the installation deviation degree of the measured coordinates of the center coordinates of each earplate through-hole through the chromatogram, as Figures 9 to 10 shown, Figure 10The example in the figure shows that the grayscale colors from light to dark represent the installation deviation degree of the measured coordinates of the center coordinates of the ear plate through hole. The darker the color, the greater the installation deviation, and the larger the number, the greater the installation deviation.

[0117] By measuring the degree of installation deviation and identifying the deviation size of each ear plate component, priority is given to performing on-site correction and adjustment on the installation of ear plates with large deviations.

[0118] The method for determining the center coordinates of the ear plate through holes of the ear plate group of the steel structure installation project based on the three-dimensional point cloud provided by the embodiment of the present invention can ensure the engineering quality and construction progress of the connection construction of the ear plate group of the steel structure installation project on the target building after the measured coordinates of the center coordinates of the ear plate through holes of the ear plate group are determined, shorten the construction period, and improve the construction efficiency. It is possible to efficiently re-measure the positioning and installation accuracy of the ear plate through holes at the construction site, and complete the component correction in time, thereby ensuring the connection quality of each ear plate in the ear plate group of a massive number of ear plates such as tens of thousands of pieces.

[0119] The present invention is not limited to the above-mentioned specific implementation modes. Obviously, the above-mentioned embodiments are only some embodiments of the embodiments of the present invention, but not all embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention. Those skilled in the art can make other levels of modifications and changes to the present invention. In this way, if these modifications and changes of the present invention fall within the scope of the claims of the present invention, the present invention is also intended to include these changes and changes.

Claims

1. A method for determining the center coordinates of the through holes of the ear plates in the ear plate group of a steel structure installation project based on 3D point clouds, characterized in that, Including: Obtaining the overall three-dimensional point cloud data of the ear plate group in the steel structure installation project; Performing a spatial transformation on the acquisition coordinate system of the overall three-dimensional point cloud data of the ear plate group in the steel structure installation project to the construction site coordinate system, and denoising and filtering the overall three-dimensional point cloud data of the ear plate group in the steel structure installation project after the spatial transformation; Through the steps of local point cloud segmentation of ear plate component nodes, ear plate plane feature fitting, ear plate point cloud attitude adjustment, ear plate through-hole contour feature extraction, and ear plate through-hole center coordinate calculation for the overall three-dimensional point cloud data of the ear plate group in the steel structure installation project after denoising and filtering, determining the measured coordinate positions of the center coordinates of each ear plate through-hole in the ear plate group of the steel structure installation project.

2. The method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on the three-dimensional point cloud according to claim 1, wherein, Also including: Checking whether the measured coordinates of the center coordinates of each ear plate through-hole are valid by comparing the differences in the measured coordinates of the center coordinates of the through-holes on both sides of each ear plate in the thickness direction in the ear plate group of the steel structure installation project after spatial transformation; Calculating the deviation between the measured coordinates of the center coordinates of the through-holes on both sides of each ear plate in the thickness direction in the effective ear plate group of the steel structure installation project after spatial transformation and the design coordinates respectively, taking the average value as the installation deviation of each ear plate for coloring and three-dimensional visualization display.

3. The method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on the three-dimensional point cloud according to claim 2, wherein Coloring the point cloud data of the measured coordinates of the center coordinates of the ear plate through-holes in a single color or in color from light to deep or from deep to light according to the degree of deviation from small to large or from large to small, so as to distinguish the degree of deviation of the measured coordinates of the center coordinates of each ear plate through-hole through the chromatogram.

4. The method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on the three-dimensional point cloud according to claim 1, wherein, The method for local point cloud segmentation of the ear plate component nodes includes: Step 311: Obtaining the BIM design model of the target building and the ear plate group in the steel structure installation project thereon, and batch extracting and exporting the set of center point pairs of the designed coordinates of the ear plate through-holes in the ear plate group of the steel structure installation project on the BIM design model; M = {(M 1L , M 1R ), (M 2L , M 2R ), …, (M iL , M iR ), …, (M nL , M nR )}, Among them, the center point pair (M iL , M iR ) represents the center points on the left and right sides of the through hole of the i-th ear plate; Step 312: Load the set of central points of the ear plate through-hole design coordinates of the ear plate group of the steel structure installation project exported from the BIM design model onto the overall three-dimensional point cloud data PC of the ear plate group of the steel structure installation project after denoising and filtering. With any point of the central point pair (M iL , M iR ) as the center, search and extract the neighboring points within the radius R in the overall three-dimensional point cloud data PC of the ear plate group of the steel structure installation project after denoising and filtering to obtain the local point cloud PC i of the ear plate component node M i , realizing the local point cloud segmentation of each ear plate component node in the overall three-dimensional point cloud data of the ear plate group of the steel structure installation project after denoising and filtering; where the search radius R = max(0.1, D), and D is the maximum size of the ear plate component node.

5. The method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on 3D point cloud according to claim 4, wherein, The method for ear plate plane feature fitting includes: Step 321: Fit the local point cloud PC of each of the ear plate component nodes M after the local point cloud segmentation of the ear plate component nodes based on the improved Random Sample Consensus (RANSAC) algorithm; specifically including: i of the local point cloud i the maximum plane feature in; specifically including: Step 321.1: Setting the initial inlier distance threshold [d] and the maximum number of iterations N, and initializing the best fitting plane; Step 321.2: Randomly extract three non-collinear points from the local point cloud PC of the gusset member node to generate a candidate plane, calculate the Euclidean distance from all data points in the local point cloud to the candidate plane, and filter out the points with Euclidean distance less than the distance threshold [d] as inliers; i Among them, calculate the Euclidean distance from all data points in the local point cloud to the candidate plane, and filter out the points with Euclidean distance less than the distance threshold [d] as inliers; Step 321.3: Counting the number of inliers of the candidate plane described in Step 321.

2. If it is greater than the number of inliers of the best fitting plane, updating the candidate plane as the best fitting plane, otherwise not updating; Step 321.4: Calculate the proportion p of the number of inliers of the current best-fitting plane in the local point cloud PC i and dynamically adjust the inlier distance threshold according to the iterative formula (1). d = d×(1 - α×p) (1); In formula (1), α is the threshold adjustment coefficient, 0 < α < 1; Step 321.5: Repeat steps 321.2 to 321.4 until the maximum number of iterations N described in step 321.1 is reached or the best-fitting plane has not been updated for 10 consecutive iterations, to determine the best-fitting plane as the local point cloud PC i with the largest plane, and output its plane feature parameters (A i , B i , C i , D i ) and the inlier set pc i .

6. The method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on 3D point cloud according to claim 5, characterized in that, The method for ear plate point cloud attitude adjustment includes: Step 331: Traverse the inlier point set pc of the maximum plane described in Step 321.5 i for the points P i (x i , y i , z i ), and project P i onto the maximum plane to obtain the projected point P i ′(x i ′, y i ′, z i ′) of the maximum plane in Formula (2) with the coordinates as follows: Step 332: Calculating and determining the rotation axis k and rotation angle θ for aligning the maximum plane described in Step 331 to the XOY plane, and constructing a rotation matrix R accordingly; Step 333: Rotate the projection point set {P i ′} according to Step 332 to adjust the point cloud attitude, and obtain the point cloud pc i ′, so as to realize the attitude adjustment of the ear plate point cloud.

7. The method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on the three-dimensional point cloud according to claim 6, characterized in that, The rotation axis k, rotation angle θ, and rotation matrix R are calculated according to formulas (3), (4), and (5) respectively: k = (A i , B i , C i ) × (0, 0, 1) = (B i , -A j , 0) (3); In formula (4), ‖(A i , B i , C i )‖ is the vector length; In Equation (5), I is a third-order identity matrix, and [k] × is the skew-symmetric matrix of k, where the skew-symmetric matrix of k is:

8. The method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on the three-dimensional point cloud according to claim 7, characterized in that, The method for ear plate through-hole contour feature extraction includes: Step 341: Calculate the minimum convex polygon of the point cloud pc i ' after the attitude adjustment of the ear plate point cloud in Step 333, and gradually optimize the outer boundary convex polygon and identify the internal hole area according to the local density information of the point cloud and the adjacent relationship of the connection points. Then, connect the hole edge points in sequence and output the inner and outer contour point sets L all ; Step 342: Use the alpha - shape algorithm to construct a rolling circle with a radius of alpha, and make it roll around the point cloud pc i ' after attitude correction described in step 333, and output the point cloud pc i ' and the data point set adjacent to the motion trajectory of the rolling circle form an outer contour point set L outer ; Step 343: For the inner and outer contour point set L described in Step 341 all and the outer contour point set L described in Step 342 outer perform a Boolean difference operation to solve for the inner contour point set L of the point cloud inner ; if there are multiple through-holes in the ear plate component, set a distance threshold d0, and perform Euclidean clustering on the inner contour point set L of the ear plate inner to obtain the contour features {L j} of each ear plate through-hole.

9. The method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on the three-dimensional point cloud according to claim 8, characterized in that, The method for calculating the center coordinates of the ear plate through-hole includes: Step 5: According to the contour feature L of the ear plate through hole j , calculate the central coordinates of the ear plate through hole, and obtain the measured coordinates of the central point of the ear plate through hole according to the inverse transformation of the projection transformation relationship described in step 333.

10. The method for determining the center coordinates of the ear plate through holes of the ear plate group in the steel structure installation project based on the three-dimensional point cloud according to claim 9, characterized in that, The method for solving the center coordinates of the ear plate through-hole includes: If the shape of the ear plate through-hole is a circle with a known diameter, then fitting the inner contour point set based on the least squares method, and solving the center coordinates of the circle as the center coordinates of the ear plate through-hole; If the shape of the through-hole of the ear plate is a circle or an ellipse with unknown geometric parameters, the inner contour point set is fitted based on the RANSAC algorithm, and the center coordinates of the inner contour point set of the circle or ellipse with unknown geometric parameters after fitting are the center coordinates of the through-hole of the ear plate; If the shape of the through-hole of the ear plate is an irregular shape, the centroid coordinates of the inner contour point set of the irregular shape are directly solved as the center coordinates of the through-hole of the ear plate.

Images