Bridge girder structure space overall deformation detection method, terminal and storage medium

Abstract

The invention provides a bridge girder structure space overall deformation detection method, a terminal and a storage medium, and relates to the technical field of bridge girder deformation detection. The method comprises the following steps: acquiring a first main beam bottom surface point cloud and a second main beam bottom surface point cloud of a to-be-detected bridge; dividing a plurality of first point cloud blocks and a plurality of second point cloud blocks from the first main beam bottom surface point cloud and the second main beam bottom surface point cloud by using a sliding window, performing plane fitting on each point cloud block through a maximum consistent minimum distance algorithm, and taking the center point of each fitting plane as a measurement point; acquiring a plurality of first measurement points and a plurality of second measurement points; and constructing a first curved surface and a second curved surface by using the plurality of first measurement points and the plurality of second measurement points respectively, and extracting Z-axis data from the first curved surface and the second curved surface respectively based on the central position of each sliding window and performing subtraction to obtain the periodic deformation of the to-be-measured bridge. The precision of deformation detection can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of bridge main beam deformation detection, and in particular to a method, terminal and storage medium for detecting the overall spatial deformation of a bridge main beam structure. Background Art

[0002] Bridge deformation is a key indicator for assessing structural safety in bridge structural health monitoring, reflecting the structural response to degradation and even damage. Extensive evidence indicates that by regularly monitoring the deformation amplitude at key points, the degree of structural stiffness degradation can be assessed and the finite element model updated. Highly accurate deformation patterns can also be used to locate and quantify structural damage. Therefore, efficient and accurate identification of structural deformation patterns—that is, identifying the deformation amplitude and form at any location—has become a key goal in bridge structural health monitoring.

[0003] Traditional bridge girder alignment monitoring methods typically rely on ground surveying, photogrammetry, and drone-based technologies. These methods have limitations, particularly in the high-altitude or complex environments of long-span suspension bridges, making it difficult to obtain high-precision, high-resolution data. Meanwhile, with the rapid development of 3D laser scanning technology, acquiring point cloud data of bridge girder structures using laser scanners has become a new research and application trend. 3D laser scanners enable high-precision, full-coverage scanning of girder structures, providing extensive 3D data on the girder's spatial deformation.

[0004] However, capturing the overall deformation of bridge girder structures using 3D laser scanning technology still faces certain challenges and problems. First, there is no systematic, automated solution for extracting point clouds from the girder structure's underside. Second, because 3D laser scanners are susceptible to environmental interference (such as wind, rain, and light) or limited by the device's accuracy during data acquisition, the resulting point cloud data often contains clustered anomalies. These anomalies may be caused by factors such as reflections and noise, further affecting the accurate restoration of the main cable's linear shape. Finally, existing research often focuses solely on planar linear deformation, lacking a method that can accurately represent overall spatial deformation. Summary of the Invention

[0005] The present application provides a method, terminal and storage medium for detecting the spatial overall deformation of a bridge main beam structure, so as to solve the problem of low accuracy in bridge deformation detection in the prior art.

[0006] In a first aspect, the present application provides a method for detecting the overall spatial deformation of a bridge main beam structure, comprising: Acquire a bottom surface point cloud of a first main beam and a bottom surface point cloud of a second main beam of the bridge to be tested, wherein the bottom surface point cloud of the first main beam is acquired later than the bottom surface point cloud of the second main beam; Using a sliding window to divide the first main beam bottom surface point cloud and the second main beam bottom surface point cloud into a plurality of first point cloud blocks and a plurality of second point cloud blocks, respectively, and performing plane fitting on each point cloud block using a maximum consistent minimum distance algorithm, using the center point of each fitting plane as a measurement point, to obtain a plurality of first measurement points and a plurality of second measurement points, wherein each first measurement point corresponds to a first point cloud block, and each second measurement point corresponds to a second point cloud block; The multiple first measurement points and the multiple second measurement points are respectively used to construct a first curved surface and a second curved surface, and based on the center position of each sliding window, Z-axis data are respectively extracted from the first curved surface and the second curved surface and subtracted to obtain the periodic deformation of the bridge to be measured.

[0007] In a second aspect, the present application provides a terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method described in the first aspect or any possible implementation of the first aspect are implemented.

[0008] In a third aspect, the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method described in the first aspect or any possible implementation of the first aspect.

[0009] The present application provides a method, terminal and storage medium for detecting the spatial overall deformation of a bridge main beam structure, which obtains a first main beam bottom surface point cloud and a second main beam bottom surface point cloud of a bridge to be tested, wherein the acquisition time of the first main beam bottom surface point cloud is later than the acquisition time of the second main beam bottom surface point cloud; uses a sliding window to divide a plurality of first point cloud blocks and a plurality of second point cloud blocks from the first main beam bottom surface point cloud and the second main beam bottom surface point cloud respectively, and performs plane fitting on each point cloud block respectively through a maximum consistent minimum distance algorithm, takes the center point of each fitting plane as a measurement point, and obtains a plurality of first measurement points and a plurality of second measurement points, wherein each first measurement point corresponds to a first point cloud block, and each second measurement point corresponds to a second point cloud block; constructs a first curved surface and a second curved surface respectively from the plurality of first measurement points and the plurality of second measurement points, and extracts Z-axis data from the first curved surface and the second curved surface based on the center position of each sliding window and makes a difference to obtain the periodic deformation of the bridge to be tested. This application obtains point cloud data of the bottom surface of the main beam at different times, and uses a sliding window and maximum consistent minimum distance algorithm for plane fitting, which can accurately extract the characteristic points of the bottom surface of the main beam, effectively reduce the influence of noise and outliers on the measurement results, and improve the accuracy of deformation detection; and this application constructs a first curved surface and a second curved surface, and extracts Z-axis data based on the center position of the sliding window for difference, which can more comprehensively reflect the overall deformation of the main beam in space, help to identify the deformation differences of the main beam at different positions, and provide an important basis for safety assessment and maintenance of the bridge. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0011] Figure 1 This is a flowchart of the implementation of the method for detecting the overall spatial deformation of the bridge main beam structure provided by the embodiment of the present application; Figure 2 This is a schematic diagram of the bridge point cloud flipped upside down provided in an embodiment of the present application; Figure 3 This is a schematic diagram of the seed point set on the bottom surface of the main beam provided in an embodiment of the present application; Figure 4 Schematic diagram of seed point features provided in an embodiment of the present application; Figure 5 This is a schematic diagram of regional normal vector features provided in an embodiment of the present application; Figure 6 This is a schematic diagram of sliding window movement provided in an embodiment of the present application; Figure 7 This is a schematic diagram of sliding window parameter settings provided by an embodiment of the present application; Figure 8 This is a schematic diagram of vertical distance provided by an embodiment of the present application; Figure 9 This is a schematic diagram of measurement points provided in an embodiment of the present application; Figure 10 This is a schematic diagram of the plane fitting process provided in an embodiment of the present application; Figure 11 This is a schematic diagram of selecting iterative retention points provided in an embodiment of the present application; Figure 12 is a schematic diagram of orthogonal distances provided in an embodiment of the present application; Figure 13 Schematic diagram of seed point selection with h orthogonal distances provided in an embodiment of the present application; Figure 14 1 is a schematic structural diagram of a device for detecting overall spatial deformation of a bridge main beam structure provided by an embodiment of the present application; Figure 15 It is a schematic diagram of a terminal provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0013] In order to make the purpose, technical solutions and advantages of this application clearer, specific embodiments will be described below with reference to the accompanying drawings.

[0014] At present, most bridge girder deformation detection methods based on three-dimensional laser scanning remain at the simple linear fitting stage, and usually assume that the data is a complete and noise-free point cloud. However, in actual scanning, there are usually a large number of clustered noise points and sparse data, which will cause errors in deformation detection by traditional methods. To address this problem, this application proposes a method for detecting the overall spatial deformation of bridge girder structures, providing more accurate and reliable technical support for bridge girder structure monitoring and health assessment. First, a method for automatically extracting the bottom surface point cloud of the main beam is proposed. This method uses a two-time cloth simulation filtering algorithm to obtain the seed points of the bottom surface of the main beam. After obtaining the seed points, the neighborhood curvature, neighborhood maximum distance and neighborhood normal vector of the neighborhood point cloud are used to further extract the complete bottom surface point cloud of the main beam using a region growing algorithm. Secondly, a spatial continuous deformation detection mode for the main beam structure is proposed. This mode combines a sliding window and a center point acquisition technology based on robust plane fitting to realize spatial deformation detection of the main beam deflection. Finally, a robust plane fitting algorithm is proposed. For the plane fitting process in each sliding window, considering the influence of a large number of abnormal noise points and sparse data, a plane fitting method based on repeated clipping of maximum consistent minimum distance is proposed.

[0015] Compared with traditional algorithms, this application achieves higher plane fitting accuracy in high-noise situations. By measuring the spatial morphology of PVC panels in the laboratory and comparing the data obtained by the total station with this application, it was verified that this application has high robustness and measurement accuracy, meeting the deformation measurement requirements of engineering practice.

[0016] Figure 1 The flowchart of the implementation method of the bridge main beam structure spatial overall deformation detection method provided in the embodiment of the present application is detailed as follows: In step 101, a bottom surface point cloud of a first main beam and a bottom surface point cloud of a second main beam of a bridge to be measured are obtained, wherein the bottom surface point cloud of the first main beam is obtained later than the bottom surface point cloud of the second main beam.

[0017] In this embodiment of the present application, a TLS scanner is used to obtain point clouds of the bottom surfaces of the first and second main beams of the bridge to be tested. The point cloud of the bottom surface of the first main beam is scanned later than the point cloud of the bottom surface of the second main beam in order to monitor the deformation of the bridge during this time interval.

[0018] In one possible implementation, obtaining the bottom surface point cloud of the first main beam and the bottom surface point cloud of the second main beam of the bridge to be measured may include: Acquire a first bridge point cloud and a second bridge point cloud of the bridge to be measured, wherein the acquisition time of the first bridge point cloud is later than the acquisition time of the second bridge point cloud; A cloth simulation filtering algorithm is used to obtain the main beam bottom surface seed point sets of the first bridge point cloud and the second bridge point cloud, and the main beam bottom surface seed point set corresponding to the first bridge point cloud is used as the first seed point set, and the main beam bottom surface seed point set corresponding to the second bridge point cloud is used as the second seed point set; The neighborhood curvature, neighborhood maximum distance and neighborhood normal vector of the first seed point set are calculated to determine the point cloud of the bottom surface of the first main beam, and the neighborhood curvature, neighborhood maximum distance and neighborhood normal vector of the second seed point set are calculated to determine the point cloud of the bottom surface of the second main beam.

[0019] Among them, Cloth Simulation Filter (CSF) is a term in computer graphics. The core idea of ​​this method is to use a piece of rigid cloth to cover the inverted point cloud surface to generate a surface similar to the ground, and then extract ground points according to the set threshold.

[0020] Since the bottom surface of the bridge main beam is located above the bottom surface below the bridge, the embodiment of the present application uses a cloth simulation filtering algorithm to obtain the point cloud of the bottom surface of the main beam of the bridge to be tested. The main process is as follows: Bridge point cloud acquisition: A first bridge point cloud and a second bridge point cloud of the bridge to be measured are acquired using a TLS scanner, and the acquisition time of the first bridge point cloud is later than the acquisition time of the second bridge point cloud.

[0021] For each set of bridge point clouds, perform the following steps: Preprocessing: Flip the bridge point cloud upside down, refer to Figure 2 As shown, Figure 2 The bridge point cloud below is directly flipped upward to obtain Figure 2 The bridge point cloud above.

[0022] Cloth mesh initialization: Construct a uniformly distributed 2D mesh based on the preset mesh resolution S. All mesh nodes are initially placed at a uniform height, which needs to be higher than the flipped point cloud data. Figure 2 , the top curve is the initial mesh. At this point, all mesh nodes are marked as "movable nodes" to allow them to participate in the subsequent dynamic simulation.

[0023] Determining the maximum height: For each mesh node, the nearest point cloud data point is searched horizontally and the elevation of that point is used as the minimum allowable height (i.e., the maximum height) for the current mesh node during its descent. This ensures that the fabric adheres closely to the point cloud surface rather than penetrating it.

[0024] Simulate fabric sinking: Under the action of gravity, let the height of all "movable nodes" drop a fixed amount H at each simulation moment dIf a node's height drops to or below the corresponding limit height during the sinking process, its height will be locked to the limit value and its status will be updated to "fixed node", indicating that the position has been fitted to the point cloud data.

[0025] Simulate cloth rebound: Considering the rebound phenomenon that may occur after the collision, for the nodes that are still in the "movable" state, their height is appropriately increased by a value H according to the preset cloth hardness parameter. u This process gives the fabric a certain degree of recovery, helping it to better adapt to the undulations of the terrain.

[0026] Iterative optimization: Repeat the "sinking" and "bounce" steps until a termination condition is met: for example, a predetermined number of iterations, M, is reached or the maximum difference in height change between two consecutive iterations falls below a certain threshold. This iterative process gradually stabilizes the fabric and makes it adhere closely to the approximate surface formed by the inverted point cloud.

[0027] Ground point filtering: Compare the elevation difference between each point in the original main beam point cloud and the corresponding grid node. When the height difference between a point and the fabric surface is less than the preset threshold H c , the point can be classified as a ground point and filtered.

[0028] Secondary CSF acquisition: After filtering the ground points, perform the above steps again, and you will get the seed point set of the main beam bottom surface. Figure 3 , Figure 3 The virtual fabric in is the seed point set of the bottom surface of the main beam.

[0029] After two CSF ​​filters, the resulting surface points are partial, but various noise points in the scene remain (e.g., billboards, roadside ancillary facilities, tree and vegetation noise, and a small amount of point cloud from bridge piers). To remove these noise points and complete the point cloud of the main beam bottom surface, this embodiment of the application develops a region growing method based on neighborhood features, specifically by calculating the neighborhood curvature, maximum neighborhood distance, and normal vector of the first and second seed point sets, respectively, to determine the corresponding first and second main beam bottom surface point clouds.

[0030] In one possible implementation, calculating the neighborhood curvature, the neighborhood maximum distance, and the neighborhood normal vector of the first seed point set to determine the bottom surface point cloud of the first main beam may include: Calculate the normal vector and curvature of each seed point in the first seed point set, and form a candidate seed point set from all seed points whose curvature is less than the curvature threshold. For each candidate seed point in the candidate seed point set, search for the number of points in the K-neighborhood point set belonging to the candidate seed point set, and determine whether the number of points is not less than the first threshold. If the number of points is not less than the first threshold, add the candidate seed point to the bottom surface point cloud of the first main beam. Calculate the Euclidean distances of each seed point in the first seed point set to all neighboring points in its K-neighborhood respectively, and for each seed point, determine whether the maximum neighborhood distance among all the Euclidean distances corresponding to the seed point is less than the maximum neighborhood radius of the seed point on the bottom surface of the bridge to be tested. If the maximum neighborhood distance among all the Euclidean distances corresponding to the seed point is less than the maximum neighborhood radius of the seed point on the bottom surface of the bridge to be tested, then add the seed point to the point cloud of the bottom surface of the first main beam; For each seed point in the first seed point set, the normal vector of the seed point is calculated, and n points within the preset search radius of the seed point are searched through a k-dimensional tree, and the distances and normal vector angles between the n points and the normal plane of the seed point are calculated respectively, and the points whose distance is less than the distance threshold and whose normal vector angle is less than the normal vector angle threshold are added to the bottom surface point cloud of the first main beam, where n is a positive integer.

[0031] In the embodiment of the present application, the neighborhood curvature, neighborhood maximum distance, and neighborhood normal vector of the first seed point set are calculated in the same way as those of the second seed point set. The calculation process of the first seed point set is taken as an example: (1) Neighborhood curvature calculation 1) Calculation of point cloud curvature value: For the seed point p, let its k-neighborhood point set be By analyzing the eigenvalues ​​and eigenvectors of the neighborhood covariance matrix, the normal vector of the seed point p is determined and curvature , the formula is as follows:

[0032]

[0033] in, 、 、 are the eigenvalues ​​of the covariance matrix, is the eigenvalue The corresponding eigenvector.

[0034] 2) Plane point screening: First, traverse the curvature of all seed points in the first seed point set and retain those with curvature below the curvature threshold The seed points constitute the candidate seed point set .

[0035] Then, for the candidate seed point set Each candidate seed point in , search its k-neighborhood point set , statistics of the k-neighborhood point set belonging to the candidate seed point set The number of points n.

[0036] If the number of points n is not less than the first threshold, then Added to the point cloud of the bottom surface of the first main beam.

[0037] The first threshold is ,in, is the plane coefficient, usually ranging from 0.5 to 1.0.

[0038] (2) Neighborhood maximum distance calculation 1) Neighborhood Euclidean distance calculation: For the seed point p, its k-neighborhood point set is defined as , seed point p to neighboring points The neighborhood Euclidean distance is Indicates that:

[0039] in, Neighborhood points The coordinate value of is the coordinate value of the seed point p.

[0040] Calculate the neighborhood Euclidean distance of all neighborhood points to obtain the neighborhood point distance set .

[0041] 2) Maximum distance constraint: For the neighborhood point distance set Sort the elements in to get the maximum neighborhood distance .like , then the seed point p satisfies the seed point condition and is added to the point cloud of the bottom surface of the first main beam. is the maximum neighborhood radius of the seed point on the bottom surface of the bridge to be tested, which is related to the number of neighborhood search points and the density of point cloud data. At the same time, in order to avoid misjudgment of narrow and long auxiliary facilities, The value must be significantly larger than the maximum neighborhood radius R of the bridge bottom.

[0042] (3) Neighborhood normal vector calculation After the neighborhood curvature and neighborhood maximum distance calculation, the seed point only has the bottom surface of the main beam, but Figure 4The seed points obtained by CSF only cover the surface layer, and a large part of the bottom surface point cloud is not covered. Therefore, the embodiment of the present application also needs to cluster points with similar attributes into the same object to achieve complete extraction of the main beam bottom surface point cloud. The main beam bottom surface point cloud is flat and smooth, that is, the change of the normal vector is also uniform; at the bottom surface boundary, the normal vector changes. Based on these characteristics, the distance threshold from the point to the normal vector is introduced. Angle threshold with normal vector The specific process is as follows: 1) Search the n nearest neighboring points of each seed point in the first seed point set through k dimensions and calculate the normal vector of each point.

[0043] 2) Within the search radius R, calculate the distance h and the normal vector angle between n points and the corresponding seed point normal plane respectively ,like Figure 5 If the distance h is less than the distance threshold , and the normal vector angle Less than the normal vector angle threshold , then add the seed point to the point cloud of the bottom surface of the first main beam.

[0044] 3) Select the points in the seed point sequence as the current seed point in turn, and return to step 2) for processing until all points in the seed point sequence are processed.

[0045] 4) Mark the initial seed point and all points in the seed point sequence as classified and group them into the same object. Clear the points in the seed point sequence to complete the region growing point cloud extraction.

[0046] In step 102, a sliding window is used to divide the first main beam bottom surface point cloud and the second main beam bottom surface point cloud into multiple first point cloud blocks and multiple second point cloud blocks, and a maximum consistent minimum distance algorithm is used to perform plane fitting on each point cloud block. The center point of each fitting plane is used as a measurement point to obtain multiple first measurement points and multiple second measurement points, wherein each first measurement point corresponds to a first point cloud block, and each second measurement point corresponds to a second point cloud block.

[0047] Normally, the axial length of a bridge beam is much larger than its cross-sectional dimensions, so the deformation mode of the bridge is usually expressed as a spatial curve. The embodiment of the present application defines that the X-axis is along the span direction of the bridge, the Y-axis is along the horizontal direction perpendicular to the span direction, and the Z-axis is along the vertical upward deformation direction. Under the action of gravity, the deformation of the bridge is expressed as a two-dimensional curve in the XZ plane; and when factors such as lateral foundation slip or uneven sunlight are considered, the deformation mode will evolve into a three-dimensional curve. However, the mode can still be decomposed into two-dimensional curves in the XY plane and the XZ plane. By calculating the deformation curves of the XY and XZ planes respectively from the point cloud, the spatial deformation mode of the bridge can be fully characterized. Among them, the XY plane deformation curve is extracted from the point cloud of the bottom or top surface of the beam, and the XZ plane deformation curve is calculated by the point cloud of the side of the beam.

[0048] Specifically, this embodiment uses a sliding window to segment the first and second main beam bottom surface point clouds into point cloud blocks with identical boundaries, namely, multiple first point cloud blocks and multiple second point cloud blocks. A maximum consistent minimum distance algorithm is then used to fit each point cloud block to suppress point cloud noise. The center point of each fitted plane is used as a measurement point, and multiple first and second measurement points are acquired through window sliding.

[0049] In one possible implementation, a sliding window is used to divide the first main beam bottom surface point cloud and the second main beam bottom surface point cloud into a plurality of first point cloud blocks and a plurality of second point cloud blocks, and a maximum consistent minimum distance algorithm is used to perform plane fitting on each point cloud block. The center point of each fitting plane is used as a measurement point to obtain the plurality of first measurement points and the plurality of second measurement points. This may include: The point cloud of the bottom surface of the first main beam and the point cloud of the bottom surface of the second main beam are registered to the same coordinate system, and a preset sliding window size and a preset step size are used to slide along the span direction and the transverse direction of the bridge to be measured to obtain a plurality of first point cloud blocks and a plurality of second point cloud blocks respectively; Calculating the center coordinates of each point cloud block, wherein the boundary of each first point cloud block is the same as the boundary of the corresponding second point cloud block; Performing plane fitting on each first point cloud block and each second point cloud block using a maximum consistent minimum distance algorithm to obtain a plurality of first fitting planes and a plurality of second fitting planes; A first measurement point of a corresponding first fitting plane is obtained based on the central coordinates of each first point cloud block, and a second measurement point of a corresponding second fitting plane is obtained based on the central coordinates of each second point cloud block.

[0050] The specific implementation process of this embodiment is as follows: (1) The TLS scanner is used to scan the bridge at different times to obtain the point cloud of the bottom surface of the first main beam and the point cloud of the bottom surface of the second main beam. The scanning time of the bottom surface cloud of the first main beam is later than that of the bottom surface cloud of the second main beam.

[0051] (2) Align the bottom surface point cloud of the first main beam and the bottom surface point cloud of the second main beam to the same coordinate system, and intercept the same target area from the two sets of registered bottom surface point clouds to obtain multiple point cloud blocks. Figure 6 ,Taking the bottom surface of the main beam as an example, the actual area needs to be manually selected according to the bridge type.

[0052] (3) Sliding window division rules: refer to Figure 7 , define the preset sliding window size (length l w , width h w ) and the preset step size s, slide along the span direction (X axis) and the transverse direction (Y axis) of the bridge to be measured.

[0053] Among them, the sliding window overlapping strategy: if the preset step size s <l w (or s <h w ), adjacent windows partially overlap, which can improve the density and continuity of measurement points.

[0054] Sliding window boundary constraint: The preset sliding window size needs to satisfy the local plane assumption, that is, the curvature radius of the point cloud within the window R>5l w .

[0055] (4) Calculate the center coordinate x of the sliding window (i.e., point cloud block) i i , as the X coordinate of the deformation measurement point i.

[0056] (5) Use the maximum consistent minimum distance algorithm to fit the point cloud blocks before and after deformation in the sliding window i, and obtain fitting plane 1 (i.e., the first fitting plane) and fitting plane 2 (i.e., the second fitting plane). Calculate the distance between fitting plane 1 and x. i The coordinates of , and find its vertical distance to fitting plane 2 ,but That is a measuring point, vertical distance Reference Figure 8 shown.

[0057] (6) Move the window to the next position (such as ), repeat steps (4) and (5) to obtain Value This is the measurement point of the deformation surface of the bridge bottom under the next sliding window. All measurement points are as follows Figure 9 shown.

[0058] In a possible implementation, plane fitting is performed on each first point cloud block and each second point cloud block using a maximum consistent minimum distance algorithm to obtain multiple first fitting planes and multiple second fitting planes, which may include: For each point cloud block, perform the following steps: Get m seed points in the point cloud block; The principal component analysis method is used to perform preliminary plane fitting on the m seed points. After obtaining the preliminary fitting plane, the orthogonal distances from the remaining seed points in the point cloud block to the preliminary fitting plane are calculated respectively. Sort all orthogonal distances in ascending order, select the seed points corresponding to the first h orthogonal distances, and use principal component analysis to perform plane fitting on the seed points corresponding to the first h orthogonal distances. After plane fitting, calculate the eigenvalues ​​of the seed points corresponding to the first h orthogonal distances respectively, and add the minimum value of the h eigenvalues ​​to the eigenvalue set; Determine whether the current number of iterations reaches the preset number of iterations; If the current number of iterations reaches the preset number of iterations, the plane corresponding to the minimum eigenvalue in the eigenvalue set is selected, and the normalized distances from all seed points in the point cloud block to the plane corresponding to the minimum eigenvalue are calculated. All seed points whose normalized distances meet the preset conditions are taken as inliers, and plane fitting is performed on all inliers using the principal component analysis method to obtain the fitting plane corresponding to the point cloud block. If the current number of iterations does not reach the preset number of iterations, the current number of iterations is increased by 1, and the process returns to the step of obtaining m seed points in the point cloud block to continue.

[0059] In bridge 3D scanning projects, point cloud data is affected by multiple factors, including sensor physical limitations, discontinuous spatial feature boundaries, environmental occlusion, multiple reflection effects, and point cloud registration errors. This inevitably results in uniform noise distribution and outlier interference. While existing mainstream plane fitting algorithms offer a certain degree of versatility, they are significantly less robust against high-proportion noise (especially outlier point clusters with spatially clustered characteristics). Therefore, for incomplete planar point cloud data containing high outliers, the present embodiment employs a maximum consistent minimum distance algorithm for plane fitting, which has significant research value in improving the accuracy of bridge structural feature extraction.

[0060] Among them, the core of the maximum consistent minimum distance algorithm is to first find the maximum consistent normal vector of the plane point cloud, and then finally determine the best fitting plane by limiting the minimum distance from other points to the plane under the normal vector constraint.

[0061] Reference Figure 10 As shown, the plane fitting process is as follows: (1) In each point cloud block, m seed points are randomly selected to fit the plane corresponding to the point cloud block.

[0062] Among them, reference Figure 11For a point cloud block, it is believed that its optimal plane should be the plane fitted by h seed points consisting of the majority of high-quality points with the highest reliability and the best identity in the local neighborhood. Generally, this embodiment sets h to between (0.5~1)k, where k is the total number of seed points in the point cloud block, that is, this embodiment determines that at least half of the seed points are close to the correct plane. In order to obtain these optimal h seed points, it is necessary to first randomly select m seed points from all seed points. M seed points are the minimum number of points required to fit the plane, that is, m=3. Using m seed points to find h seed points can greatly reduce the iteration time because m seed points are much fewer than h seed points.

[0063] (2) For the m seed points above, use the principal component analysis method to perform preliminary plane fitting on the m seed points. After obtaining the preliminary fitting plane, calculate the orthogonal distance OD from the remaining seed points in the point cloud block to the preliminary fitting plane. Figure 12 As shown, all orthogonal distances are sorted in order from small to large: .

[0064] (3) Reference Figure 13 , select the seed points corresponding to the first h orthogonal distances OD, use the principal component analysis method to perform plane fitting on the h seed points, and after plane fitting, calculate the eigenvalues ​​of the h seed points respectively, and take the minimum value Add to feature collection middle.

[0065] (4) Determine whether the preset number of iterations has been reached , if reached, go to (5); otherwise, , return to step (1).

[0066] (5) From the feature set Find the plane corresponding to the minimum eigenvalue, calculate the normalized distances from all seed points in the point cloud block to the plane, and take all seed points whose normalized distances meet the preset conditions as inliers. Use the principal component analysis method to perform plane fitting on all inliers again to obtain the fitting plane corresponding to the point cloud block.

[0067] In a possible implementation, after obtaining m seed points in the point cloud block, the method may further include: Take the point set consisting of m seed points as the first point set, calculate the rank of the first point set, and determine whether the rank of the first point set is less than m; If the rank of the first point set is less than m, randomly add a preset number of seed points to the first point set and return to the step of obtaining m seed points in the point cloud block to continue execution; If the rank of the first point set is equal to m, a preliminary plane fitting step is performed on the m seed points using principal component analysis.

[0068] Optionally, before selecting m seed points for preliminary plane fitting, it is also necessary to determine whether the rank of the first point set composed of m seed points is less than m. If it is less than m, more seed points are randomly added to the first point set and re-judgment is performed; if it is equal to m, preliminary plane fitting is performed using m seed points.

[0069] In one possible implementation, the calculation process of the preset number of iterations is: Monte Carlo probability simulation is used to calculate the preset number of iterations.

[0070] Optionally, randomly select m seed points and iterate continuously, with a preset number of iterations Introducing Monte Carlo probability simulation, the calculation formula is:

[0071] in, is the outlier rate of the point cloud block, which can be set by the engineer; For containing Find at least one outlier-free point in the outlier data The expected probability of a seed point: .

[0072] For example, in the embodiment of the present application, m=3 and It should be noted that higher The value will increase the number of iterations, but it can increase the probability of the subset having no outliers, so the number of iterations is essentially a trade-off between accuracy and efficiency. It is unknown and underestimated will be affected by the masking effect and overestimate This will result in the real points being submerged. Data experience shows that most points (>50%) in a local neighborhood are internal points, so the actual data is assumed to be Through the above settings, we can get that when m=3, , Under these conditions, the number of iterations 68 times.

[0073] In a possible implementation, the preset condition is that the distance is not greater than the second threshold, and all seed points whose normalized distances meet the preset condition are considered as inliers, which may include: For each seed point, determine whether the normalized distance corresponding to the seed point is not greater than the second threshold. If the normalized distance corresponding to the seed point is not greater than the second threshold, the seed point is determined as an inner point; if the normalized distance corresponding to the seed point is greater than the second threshold, the seed point is determined as an outer point and deleted from all seed points in the point cloud block.

[0074] Among them, for the maximum consistent minimum distance algorithm, the previous algorithm used the mean and standard deviation to calculate the standardized distance from the point to the fitting plane in the inlier screening stage, but the mean and standard deviation themselves are very sensitive to outliers. Therefore, the embodiment of the present application also introduces the median (med) and median absolute deviation (MAD) in this step, so that even if there are a large number of outliers in the data, the median will not be biased, and the inlier screening is more reliable.

[0075] After each iteration, h seed points are selected based on the minimum orthogonal distance OD of the plane fitted by the m seed points, and the eigenvalues ​​of the h seed points are calculated. The orthogonal distance OD is calculated by the following formula:

[0076] in, is the seed point, is the projection point from the seed point to the preliminary fitting plane, is the unit normal vector of the initial fitting plane.

[0077] Theoretically, the smallest eigenvalue The corresponding plane has the greatest surface consistency (i.e., the variation along the normal direction is minimal). However, there are still some deviation points in the spatial plane at this time. In this embodiment of the application, this plane is first used as a preliminary plane model, i.e., the best candidate consistent subset (MCS). The plane model at this location is used to recalculate the orthogonal distance OD of all points, and the median med and median absolute deviation MAD of the orthogonal distance OD are introduced. The median absolute deviation MAD is:

[0078] in, is a correction coefficient that can be used to improve the consistency of the estimate and can be taken as 1.4826.

[0079] Calculate the normalized distance for each seed point as:

[0080] in, To prevent division by zero, Set the threshold value, that is, the second threshold value, to 2.5. Points smaller than the second threshold are determined as inliers, and the filtered inlier data are fitted again using the principal component analysis method to obtain the final fitting plane.

[0081] In step 103, a first curved surface and a second curved surface are constructed using multiple first measurement points and multiple second measurement points respectively, and based on the center position of each sliding window, Z-axis data are extracted from the first curved surface and the second curved surface respectively and subtracted to obtain the periodic deformation of the bridge to be measured.

[0082] In an embodiment of the present application, a first curved surface is constructed using multiple first measurement points, and a second curved surface is constructed using multiple second measurement points. Then, through a sliding window, the Z-axis data at the center position of the sliding window with the same boundary is extracted from the first curved surface and the second curved surface respectively, and the difference is made to determine the deformation of the bridge to be measured.

[0083] It should be noted that when the deformation is the normal vector distance between two planes, it is applicable to beams and tower components with uniform cross-sections; for components with variable cross-sections, the normal distance is Converted to vertical deformation ,Right now:

[0084] in, is the angle between the surface normal and the deformation direction at the measuring point.

[0085] The present application provides a method for detecting the spatial overall deformation of a bridge main beam structure, which obtains a first main beam bottom surface point cloud and a second main beam bottom surface point cloud of a bridge to be tested, wherein the acquisition time of the first main beam bottom surface point cloud is later than the acquisition time of the second main beam bottom surface point cloud; uses a sliding window to divide a plurality of first point cloud blocks and a plurality of second point cloud blocks from the first main beam bottom surface point cloud and the second main beam bottom surface point cloud, respectively, and performs plane fitting on each point cloud block respectively through a maximum consistent minimum distance algorithm, takes the center point of each fitting plane as a measurement point, and obtains a plurality of first measurement points and a plurality of second measurement points, wherein each first measurement point corresponds to a first point cloud block, and each second measurement point corresponds to a second point cloud block; constructs a first curved surface and a second curved surface respectively from the plurality of first measurement points and the plurality of second measurement points, and extracts Z-axis data from the first curved surface and the second curved surface based on the center position of each sliding window and makes a difference to obtain the periodic deformation of the bridge to be tested. This application obtains point cloud data of the bottom surface of the main beam at different times, and uses a sliding window and maximum consistent minimum distance algorithm for plane fitting, which can accurately extract the characteristic points of the bottom surface of the main beam, effectively reduce the influence of noise and outliers on the measurement results, and improve the accuracy of deformation detection; and this application constructs a first curved surface and a second curved surface, and extracts Z-axis data based on the center position of the sliding window for difference, which can more comprehensively reflect the overall deformation of the main beam in space, help to identify the deformation differences of the main beam at different positions, and provide an important basis for safety assessment and maintenance of the bridge.

[0086] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0087] The following are device embodiments of the present application. For details not fully described therein, please refer to the corresponding method embodiments described above.

[0088] Figure 14 The following is a schematic diagram of the structure of a device for detecting the overall spatial deformation of a bridge main beam structure provided by an embodiment of the present application. For ease of explanation, only the parts related to the embodiment of the present application are shown, which are described in detail as follows: like Figure 14 As shown, the bridge main beam structure spatial overall deformation detection device 14 includes: An acquisition module 141 is configured to acquire a bottom surface point cloud of a first main beam and a bottom surface point cloud of a second main beam of the bridge to be tested, wherein the bottom surface point cloud of the first main beam is acquired later than the bottom surface point cloud of the second main beam; a division and fitting module 142 for dividing the first main beam bottom surface point cloud and the second main beam bottom surface point cloud into a plurality of first point cloud blocks and a plurality of second point cloud blocks using a sliding window, performing plane fitting on each point cloud block using a maximum consistent minimum distance algorithm, taking the center point of each fitting plane as a measurement point, and obtaining a plurality of first measurement points and a plurality of second measurement points, wherein each first measurement point corresponds to a first point cloud block, and each second measurement point corresponds to a second point cloud block; The deformation determination module 143 is used to construct a first curved surface and a second curved surface using multiple first measurement points and multiple second measurement points respectively, and based on the center position of each sliding window, extract Z-axis data from the first curved surface and the second curved surface respectively and make a difference to obtain the periodic deformation of the bridge to be measured.

[0089] The present application provides a device for detecting the spatial overall deformation of a bridge main beam structure, which obtains a first main beam bottom surface point cloud and a second main beam bottom surface point cloud of a bridge to be tested, wherein the acquisition time of the first main beam bottom surface point cloud is later than the acquisition time of the second main beam bottom surface point cloud; uses a sliding window to divide a plurality of first point cloud blocks and a plurality of second point cloud blocks from the first main beam bottom surface point cloud and the second main beam bottom surface point cloud, respectively, and performs plane fitting on each point cloud block respectively through a maximum consistent minimum distance algorithm, takes the center point of each fitting plane as a measurement point, and obtains a plurality of first measurement points and a plurality of second measurement points, wherein each first measurement point corresponds to a first point cloud block, and each second measurement point corresponds to a second point cloud block; constructs a first curved surface and a second curved surface respectively using the plurality of first measurement points and the plurality of second measurement points, and extracts Z-axis data from the first curved surface and the second curved surface based on the center position of each sliding window and makes a difference to obtain the periodic deformation of the bridge to be tested. This application obtains point cloud data of the bottom surface of the main beam at different times, and uses a sliding window and maximum consistent minimum distance algorithm for plane fitting, which can accurately extract the characteristic points of the bottom surface of the main beam, effectively reduce the influence of noise and outliers on the measurement results, and improve the accuracy of deformation detection; and this application constructs a first curved surface and a second curved surface, and extracts Z-axis data based on the center position of the sliding window for difference, which can more comprehensively reflect the overall deformation of the main beam in space, help to identify the deformation differences of the main beam at different positions, and provide an important basis for safety assessment and maintenance of the bridge.

[0090] In a possible implementation, the acquisition module may be specifically used to: Acquire a first bridge point cloud and a second bridge point cloud of the bridge to be measured, wherein the acquisition time of the first bridge point cloud is later than the acquisition time of the second bridge point cloud; A cloth simulation filtering algorithm is used to obtain the main beam bottom surface seed point sets of the first bridge point cloud and the second bridge point cloud, and the main beam bottom surface seed point set corresponding to the first bridge point cloud is used as the first seed point set, and the main beam bottom surface seed point set corresponding to the second bridge point cloud is used as the second seed point set; The neighborhood curvature, neighborhood maximum distance and neighborhood normal vector of the first seed point set are calculated to determine the point cloud of the bottom surface of the first main beam, and the neighborhood curvature, neighborhood maximum distance and neighborhood normal vector of the second seed point set are calculated to determine the point cloud of the bottom surface of the second main beam.

[0091] In one possible implementation, the acquisition module may be used to: Calculate the normal vector and curvature of each seed point in the first seed point set, and form a candidate seed point set from all seed points whose curvature is less than the curvature threshold. For each candidate seed point in the candidate seed point set, search for the number of points in the K-neighborhood point set belonging to the candidate seed point set, and determine whether the number of points is not less than the first threshold. If the number of points is not less than the first threshold, add the candidate seed point to the bottom surface point cloud of the first main beam. Calculate the Euclidean distances of each seed point in the first seed point set to all neighboring points in its K-neighborhood respectively, and for each seed point, determine whether the maximum neighborhood distance among all the Euclidean distances corresponding to the seed point is less than the maximum neighborhood radius of the seed point on the bottom surface of the bridge to be tested. If the maximum neighborhood distance among all the Euclidean distances corresponding to the seed point is less than the maximum neighborhood radius of the seed point on the bottom surface of the bridge to be tested, then add the seed point to the point cloud of the bottom surface of the first main beam; For each seed point in the first seed point set, the normal vector of the seed point is calculated, and n points within the preset search radius of the seed point are searched through a k-dimensional tree, and the distances and normal vector angles between the n points and the normal plane of the seed point are calculated respectively, and the points whose distance is less than the distance threshold and whose normal vector angle is less than the normal vector angle threshold are added to the bottom surface point cloud of the first main beam, where n is a positive integer.

[0092] In one possible implementation, the partitioning and fitting module may be specifically used to: The point cloud of the bottom surface of the first main beam and the point cloud of the bottom surface of the second main beam are registered to the same coordinate system, and a preset sliding window size and a preset step size are used to slide along the span direction and the transverse direction of the bridge to be measured to obtain a plurality of first point cloud blocks and a plurality of second point cloud blocks respectively; Calculating the center coordinates of each point cloud block, wherein the boundary of each first point cloud block is the same as the boundary of the corresponding second point cloud block; Performing plane fitting on each first point cloud block and each second point cloud block using a maximum consistent minimum distance algorithm to obtain a plurality of first fitting planes and a plurality of second fitting planes; A first measurement point of a corresponding first fitting plane is obtained based on the central coordinates of each first point cloud block, and a second measurement point of a corresponding second fitting plane is obtained based on the central coordinates of each second point cloud block.

[0093] In one possible implementation, the partitioning and fitting module can be used to: For each point cloud block, perform the following steps: Get m seed points in the point cloud block; The principal component analysis method is used to perform preliminary plane fitting on the m seed points. After obtaining the preliminary fitting plane, the orthogonal distances from the remaining seed points in the point cloud block to the preliminary fitting plane are calculated respectively. Sort all orthogonal distances in ascending order, select the seed points corresponding to the first h orthogonal distances, and use principal component analysis to perform plane fitting on the seed points corresponding to the first h orthogonal distances. After plane fitting, calculate the eigenvalues ​​of the seed points corresponding to the first h orthogonal distances respectively, and add the minimum value of the h eigenvalues ​​to the eigenvalue set; Determine whether the current number of iterations reaches the preset number of iterations; If the current number of iterations reaches the preset number of iterations, the plane corresponding to the minimum eigenvalue in the eigenvalue set is selected, and the normalized distances from all seed points in the point cloud block to the plane corresponding to the minimum eigenvalue are calculated. All seed points whose normalized distances meet the preset conditions are taken as inliers, and plane fitting is performed on all inliers using the principal component analysis method to obtain the fitting plane corresponding to the point cloud block. If the current number of iterations does not reach the preset number of iterations, the current number of iterations is increased by 1, and the process returns to the step of obtaining m seed points in the point cloud block to continue.

[0094] In a possible implementation, the device may further include a point rank determination module, which may be configured to: Take the point set consisting of m seed points as the first point set, calculate the rank of the first point set, and determine whether the rank of the first point set is less than m; If the rank of the first point set is less than m, randomly add a preset number of seed points to the first point set and return to the step of obtaining m seed points in the point cloud block to continue execution; If the rank of the first point set is equal to m, a preliminary plane fitting step is performed on the m seed points using principal component analysis.

[0095] In a possible implementation, the preset condition is that the value is not greater than the second threshold value, and the division and fitting module can be used to: For each seed point, determine whether the normalized distance corresponding to the seed point is not greater than the second threshold. If the normalized distance corresponding to the seed point is not greater than the second threshold, the seed point is determined as an inner point; if the normalized distance corresponding to the seed point is greater than the second threshold, the seed point is determined as an outer point and deleted from all seed points in the point cloud block.

[0096] In one possible implementation, the calculation process of the preset number of iterations is: Monte Carlo probability simulation is used to calculate the preset number of iterations.

[0097] Figure 15 Schematic diagram of the terminal provided in the embodiment of the present application. Figure 15 As shown, the terminal 15 of this embodiment includes: a processor 150, a memory 151, and a computer program 152 stored in the memory 151 and executable on the processor 150. When the processor 150 executes the computer program 152, the steps in the above-mentioned embodiments of the method for detecting the spatial overall deformation of the bridge main beam structure are implemented, such as Figure 1 Alternatively, when the processor 150 executes the computer program 152, the functions of the modules / units in the above-mentioned device embodiments are realized, for example Figure 14 The functions of each module are shown.

[0098] Exemplarily, the computer program 152 may be divided into one or more modules / units, which are stored in the memory 151 and executed by the processor 150 to complete the present application. The one or more modules / units may be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program 152 in the terminal 15. For example, the computer program 152 may be divided into Figure 14 The modules shown.

[0099] The terminal 15 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal 15 may include, but is not limited to, a processor 150 and a memory 151. Those skilled in the art will understand that Figure 15 It is only an example of terminal 15 and does not constitute a limitation on terminal 15. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

[0100] The processor 150 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0101] The memory 151 can be an internal storage unit of the terminal 15, such as a hard drive or memory of the terminal 15. The memory 151 can also be an external storage device of the terminal 15, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped with the terminal 15. Furthermore, the memory 151 can include both the internal storage unit of the terminal 15 and an external storage device. The memory 151 is used to store the computer program and other programs and data required by the terminal. The memory 151 can also be used to temporarily store data that has been output or is about to be output.

[0102] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0103] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0104] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0105] In the embodiments provided in this application, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For example, the division of the modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0106] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0107] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0108] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application can also implement all or part of the processes in the above-mentioned method embodiments by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned bridge main beam structure spatial integral deformation detection method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium.

[0109] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for detecting the overall spatial deformation of a bridge main beam structure, characterized in that: include: Acquire a bottom surface point cloud of a first main beam and a bottom surface point cloud of a second main beam of the bridge to be tested, wherein the bottom surface point cloud of the first main beam is acquired later than the bottom surface point cloud of the second main beam; Using a sliding window to divide the first main beam bottom surface point cloud and the second main beam bottom surface point cloud into a plurality of first point cloud blocks and a plurality of second point cloud blocks, respectively, and performing plane fitting on each point cloud block using a maximum consistent minimum distance algorithm, using the center point of each fitting plane as a measurement point, to obtain a plurality of first measurement points and a plurality of second measurement points, wherein each first measurement point corresponds to a first point cloud block, and each second measurement point corresponds to a second point cloud block; The multiple first measurement points and the multiple second measurement points are respectively used to construct a first curved surface and a second curved surface, and based on the center position of each sliding window, Z-axis data are respectively extracted from the first curved surface and the second curved surface and subtracted to obtain the periodic deformation of the bridge to be measured.

2. The method for detecting spatial overall deformation of a bridge main beam structure according to claim 1, characterized in that: The method of obtaining the bottom surface point cloud of the first main beam and the bottom surface point cloud of the second main beam of the bridge to be measured comprises: Acquire a first bridge point cloud and a second bridge point cloud of the bridge to be measured, wherein the acquisition time of the first bridge point cloud is later than the acquisition time of the second bridge point cloud; Using a cloth simulation filtering algorithm to obtain main beam bottom surface seed point sets of the first bridge point cloud and the second bridge point cloud, and using the main beam bottom surface seed point set corresponding to the first bridge point cloud as a first seed point set, and using the main beam bottom surface seed point set corresponding to the second bridge point cloud as a second seed point set; Calculate the neighborhood curvature, neighborhood maximum distance and neighborhood normal vector of the first seed point set to determine the first main beam bottom surface point cloud, and calculate the neighborhood curvature, neighborhood maximum distance and neighborhood normal vector of the second seed point set to determine the second main beam bottom surface point cloud.

3. The method for detecting spatial overall deformation of a bridge main beam structure according to claim 2, characterized in that: The calculating the neighborhood curvature, the neighborhood maximum distance, and the neighborhood normal vector of the first seed point set to determine the first main beam bottom surface point cloud includes: Calculating the normal vector and curvature of each seed point in the first seed point set, forming a candidate seed point set from all seed points whose curvature is less than a curvature threshold, and for each candidate seed point in the candidate seed point set, searching for the number of points in the K-neighborhood point set belonging to the candidate seed point set, determining whether the number of points is not less than a first threshold, and if the number of points is not less than the first threshold, adding the candidate seed point to the first main beam bottom surface point cloud; Calculating the Euclidean distances of each seed point in the first seed point set to all neighboring points in its K-neighborhood respectively, and for each seed point, determining whether the maximum neighborhood distance among all the Euclidean distances corresponding to the seed point is less than the maximum neighborhood radius of the seed point on the bottom surface of the bridge to be measured; if the maximum neighborhood distance among all the Euclidean distances corresponding to the seed point is less than the maximum neighborhood radius of the seed point on the bottom surface of the bridge to be measured, then adding the seed point to the first main beam bottom surface point cloud; For each seed point in the first seed point set, the normal vector of the seed point is calculated, and n points within a preset search radius of the seed point are searched through a k-dimensional tree, and the distances and normal vector angles between the n points and the normal plane of the seed point are calculated respectively, and points whose distances are less than a distance threshold and whose normal vector angles are less than a normal vector angle threshold are added to the first main beam bottom surface point cloud, where n is a positive integer.

4. The method for detecting spatial overall deformation of a bridge main beam structure according to claim 1, characterized in that: The method comprises the following steps: dividing the first main beam bottom surface point cloud and the second main beam bottom surface point cloud into a plurality of first point cloud blocks and a plurality of second point cloud blocks respectively from the first main beam bottom surface point cloud and the second main beam bottom surface point cloud respectively from the first main beam bottom surface point cloud and performing plane fitting on each point cloud block respectively by using a maximum consistent minimum distance algorithm, taking the center point of each fitting plane as a measurement point, and obtaining the plurality of first measurement points and the plurality of second measurement points. Registering the first main beam bottom surface point cloud and the second main beam bottom surface point cloud to the same coordinate system, and sliding along the span direction and the transverse direction of the bridge to be measured with a preset sliding window size and a preset step size to obtain a plurality of first point cloud blocks and a plurality of second point cloud blocks respectively; Calculating the center coordinates of each point cloud block, wherein the boundary of each first point cloud block is the same as the boundary of the corresponding second point cloud block; Performing plane fitting on each first point cloud block and each second point cloud block using the maximum consistent minimum distance algorithm to obtain a plurality of first fitting planes and a plurality of second fitting planes; A first measurement point of a corresponding first fitting plane is obtained based on the central coordinates of each first point cloud block, and a second measurement point of a corresponding second fitting plane is obtained based on the central coordinates of each second point cloud block.

5. The method for detecting spatial overall deformation of a bridge main beam structure according to claim 4, characterized in that: The using the maximum consistent minimum distance algorithm to perform plane fitting on each first point cloud block and each second point cloud block respectively to obtain multiple first fitting planes and multiple second fitting planes includes: For each point cloud block, perform the following steps: Get m seed points in the point cloud block; Performing a preliminary plane fitting on the m seed points using the principal component analysis method, and after obtaining the preliminary fitting plane, respectively calculating the orthogonal distances between the remaining seed points in the point cloud block and the preliminary fitting plane; Sort all orthogonal distances in ascending order, select seed points corresponding to the first h orthogonal distances, and perform plane fitting on the seed points corresponding to the first h orthogonal distances using the principal component analysis method. After the plane fitting, calculate the eigenvalues ​​of the seed points corresponding to the first h orthogonal distances, and add the minimum value of the h eigenvalues ​​to the eigenvalue set; Determine whether the current number of iterations reaches the preset number of iterations; If the current number of iterations reaches the preset number of iterations, the plane corresponding to the minimum eigenvalue in the eigenvalue set is selected, and the normalized distances from all seed points in the point cloud block to the plane corresponding to the minimum eigenvalue are calculated. All seed points whose normalized distances meet the preset conditions are taken as inliers, and plane fitting is performed on all inliers using the principal component analysis method to obtain the fitting plane corresponding to the point cloud block; If the current number of iterations does not reach the preset number of iterations, the current number of iterations is increased by 1, and the process returns to the step of obtaining m seed points in the point cloud block and continues to execute.

6. The method for detecting spatial overall deformation of a bridge main beam structure according to claim 5, characterized in that: After obtaining the m seed points in the point cloud block, the method further includes: Taking a point set consisting of m seed points as a first point set, calculating a rank of the first point set, and determining whether the rank of the first point set is less than m; If the rank of the first point set is less than m, randomly adding a preset number of seed points to the first point set, and returning to the step of obtaining m seed points in the point cloud block to continue; If the rank of the first point set is equal to m, the step of performing preliminary plane fitting on the m seed points using the principal component analysis method is performed.

7. The method for detecting spatial overall deformation of a bridge main beam structure according to claim 5, characterized in that: The preset condition is that the distance is not greater than the second threshold, and taking all seed points whose normalized distance meets the preset condition as inliers includes: For each seed point, determine whether the normalized distance corresponding to the seed point is not greater than the second threshold. If the normalized distance corresponding to the seed point is not greater than the second threshold, the seed point is determined as an internal point; if the normalized distance corresponding to the seed point is greater than the second threshold, the seed point is determined as an external point and deleted from all seed points in the point cloud block.

8. The method for detecting spatial overall deformation of a bridge main beam structure according to claim 5, characterized in that: The calculation process of the preset number of iterations is: The preset number of iterations is calculated using Monte Carlo probability simulation.

9. A terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method for detecting the spatial overall deformation of the bridge main beam structure as described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for detecting the spatial overall deformation of the bridge main beam structure as described in any one of claims 1 to 8 are implemented.

Images