Virtual pre-assembly method for bridge closure section

Abstract

The invention discloses a virtual pre-assembly method for a bridge closure section. The method comprises the following steps: acquiring three-dimensional point cloud data of the closure section and a closure gap; carrying out closure section connection section area data extraction and closure gap connection section area data extraction based on a bisection method, and obtaining connection end face area data; extracting connection end face point cloud data from the connection end face region; extracting outer boundaries of the connecting end faces and fitting feature angular points to obtain all key angular points on each connecting end face, and finally obtaining a feature angular point data set of the closure section and the closure gap; and feature point three-dimensional model construction and virtual pre-assembly: establishing a feature point three-dimensional model based on the feature angular points of the closure section and the closure gap, simulating the final state of hoisting butt joint by using a virtual pre-assembly method, and calculating the matching and cutting amount of the closure section. According to the method, the cutting amount of each feature point position can be automatically calculated, and the precision and efficiency of closure construction are greatly improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of steel structure assembly, and particularly relates to a virtual pre-assembly method for a bridge closure section. BACKGROUND

[0002] The closure construction is the last key procedure for completing the main structure of a bridge, and the installation quality thereof is directly related to the overall structural performance, service life and safety of later operation of the bridge; In order to ensure that the closure section can be smoothly and accurately installed, a certain length allowance is usually reserved on both sides of the closure section during the factory production stage, so as to prevent serious rework in the closure process due to insufficient size. After the closure state is locked, the detection personnel measures the three-dimensional information of the feature points of the closure opening and the closure section through a total station, then imports the data into a computer-aided software to establish a three-dimensional model, and manually aligns the feature point pairs, so as to determine the cutting amount of each feature point of the closure section, and guide the cutting and hoisting operation of the closure section on site according to the cutting amount information. At present, in order to improve the accuracy, an entity pre-assembly scheme is commonly used in steel structure construction, which verifies the installability of the closure section and the closure opening through temporary facilities set up on site, but this method needs additional investment in mechanical equipment, site resources and a large number of manpower, and not only has a complex detection process, long time consumption and high cost, but also has certain construction safety risks. SUMMARY

[0003] The application aims to overcome the defects of the prior art and provides a virtual pre-assembly method for a bridge closure section, which obtains the built state point cloud data of the closure section and the closure opening based on a three-dimensional laser scanner, introduces a connection interface data extraction process based on the dichotomy idea, and combines principal component analysis, neighbor searching and boundary feature point fitting technology to automatically establish the spatial feature point model of the bridge closure section and the closure opening.

[0004] The application achieves the above-mentioned purpose through the following technical scheme: A virtual pre-assembly method for a bridge closure section, which obtains three-dimensional point cloud data of the closure section and the closure opening, and pre-processes the obtained three-dimensional point cloud data; The connection section connection section area data extraction and the closure gap connection section area data extraction are performed based on dichotomy, a cutting result reservation mode is determined according to a target area direction, and connection end surface area data is obtained; The connection end surface point cloud data is extracted from the connection end surface area by combining the field search with the projection surface accurate connection end surface data extraction; The connection end surface outer boundary extraction and feature corner fitting are performed based on a point cloud boundary extraction algorithm, a clustering algorithm and a straight line fitting algorithm, all key corner points on each connection end surface are obtained, and finally the feature corner point data set of the closure gap and the closure gap is obtained; The feature point three-dimensional model is constructed and virtual pre-assembly is performed, the feature point three-dimensional model is established based on the feature corner points of the closure gap and the closure gap, and the final state of hoisting and docking is simulated by using the virtual pre-assembly method, and the closure gap matching amount is calculated.

[0005] In one embodiment, the three-dimensional point cloud data of the closure gap and the closure gap is collected by using multiple land three-dimensional laser scanners, and the point data of the connection end surface area is encrypted during the collection stage, and the point cloud data obtained by adjacent scanners is overlapped.

[0006] In one embodiment, the obtained three-dimensional point cloud data is preprocessed, including: The three-dimensional point cloud data is registered, the environmental noise point cloud is removed, and the coordinates are unified.

[0007] In one embodiment, the closure gap connection section area data extraction is performed based on dichotomy, including: The closure gap point cloud coordinates are centralized, the point cloud centroid is calculated, and the centroid O is taken as the coordinate origin; An initial cutting surface is determined, principal component analysis is performed on the closure gap point cloud data, then a principal plane is selected and translated through the point cloud centroid, so that the initial cutting surface is determined ; The first segmentation is performed, the initial cutting surface is taken as the division basis, and the closure gap point cloud is divided into and two parts of point clouds; The second segmentation is performed, first, the new centroid coordinates of and two point clouds are calculated respectively, and are recorded as and , and the vectors and are calculated, principal component analysis is performed on and two point clouds respectively, and the third principal plane passing through the centroids of and is taken as the segmentation surface of the second point cloud segmentation and , and then a second segmentation is performed to obtain four sub-point cloud data , , and , the order of which represents the relative position between the sub-point clouds, and the centroid coordinates of the four sub-point cloud data are , , and , respectively, the centroid vectors , , and are calculated, respectively, and then the four vectors are respectively subjected to vector dot product with the corresponding vectors and , and the point cloud data corresponding to the centroid with a positive calculation result is taken as the data to be reserved; The second cutting step is iterated i times.

[0008] In an embodiment, the data extraction of the connection section area is performed based on bisection, including: Centering the connection section point cloud coordinates, calculating the point cloud centroid, and taking the centroid O as the coordinate origin; Determining the initial cutting plane, performing principal component analysis on the connection section point cloud data, then selecting a principal plane and translating it through the point cloud centroid to determine the initial cutting plane ; Performing a first segmentation to divide the connection section point cloud into and two parts of point cloud based on the initial cutting plane ; Performing a second segmentation, first calculating the new centroid coordinates of and two point clouds, respectively, and recording them as and , and calculating the vectors and , performing principal component analysis on and two point clouds, respectively, and taking the third principal plane passing through the centroids of and as the cutting plane and for the second point cloud segmentation, and then performing a second segmentation to obtain four sub-point cloud data , , and , the order of which represents the relative position between the sub-point clouds, and the centroid coordinates of the four sub-point cloud data are , , as well as Calculate the centroid vector respectively , , as well as Then these four vectors are respectively compared with their corresponding vectors. and Perform vector dot product, and retain the point cloud data corresponding to the centroids with positive results as the retained data. Iterate through the second cutting step i times.

[0009] In one implementation, extracting connection end face point cloud data from the connection end face region includes: Calculate the point cloud obtained after the (i-1)th cut from the retained point cloud data. and The corresponding centroids and The centroids of the point clouds Li-1 and Ri-1 obtained by the i-th cut and The resulting vector and ; The transformation from the cutting plane to the projection plane involves moving the planes PLi and PRi used in the i-th iteration along the vector... and The direction is translated by a factor of ten, thus constructing two projection planes. and ; Target point cloud projection: Project point clouds Li-1 and Ri-1 onto the projection plane respectively. and Above, the projected point cloud is obtained. and ; Extract the nearest point from the source point cloud, and traverse the point cloud respectively. and Point data, and use the nearest neighbor algorithm to search in the point cloud. and The K points that are closest to the projection point on the projection plane.

[0010] In one embodiment, the method further includes noise reduction processing on the extracted point cloud data of the closure segment and the connection end face of the closure port.

[0011] In one embodiment, the noise reduction process includes the following steps: The extracted data is processed by a duplicate point removal algorithm based on a distance threshold; The clustering algorithm is used to cluster each connection end surface data of the extracted closure section and closure gap, and then the connection end surface clustering data belonging to the closure section is screened out.

[0012] In one embodiment, the steps of extracting feature corner points from the closure section data and the closure gap data are as follows: The point cloud boundary extraction algorithm is used to extract the outer boundary point cloud of each connection interface data of the closure section and the closure gap after noise reduction processing. The clustering algorithm is used to cluster and identify the straight line data in the connection interface data, the straight line fitting algorithm is used to fit each straight line point cloud cluster, and a k value is set to filter the point cloud clusters that do not meet the requirements. The starting point and the end point of each connection interface outer boundary data fitting straight line are initialized. The starting point and the end point of each straight line are rearranged in sequence. According to the obtained sorting result, the nearest two points of each straight line and the next straight line are calculated, and then the midpoint of the two points is calculated as the intersection point of the two straight lines, until all the straight line intersection point three-dimensional coordinate data are obtained.

[0013] In one embodiment, the steps of establishing a feature point three-dimensional model and using a virtual pre-assembly method to simulate hoisting and docking are as follows: The feature point model of the two connection interfaces of the closure section and the feature point model of the two connection interfaces of the closure gap are established respectively. Based on the registration algorithm, the closure section feature point data is registered into the closure gap data while keeping the position of the closure gap data unchanged, and the loss function is:

[0014] Wherein, n is the total number of closure section feature corner points, And In each i=1 to n range one by one, Indicates the Euclidean norm; The projection length of each point pair along the corresponding closure gap connection surface normal vector direction is calculated and used as the cutting amount of the closure section.

[0015] The beneficial effects of the present application are as follows: The application combines virtual pre-assembly technology and high-precision feature point extraction algorithm, obtains the built state point cloud data of the closure section and closure opening based on a three-dimensional laser scanner, introduces a connection interface data extraction process based on the dichotomy idea, and combines principal component analysis, nearest neighbor search and boundary feature point fitting technology to realize automatic establishment of the spatial feature point model of the bridge closure section and closure opening, and then through the virtual pre-assembly technology, the cutting amount at each feature point position is automatically calculated according to the feature point model registration alignment, that is, the problems of low efficiency, high error and high dependence on experience existing in the traditional total station + manual alignment mode are effectively solved, the precision and efficiency of the bridge closure section cutting construction are significantly improved, the precision and timeliness in the construction process are significantly improved, and a reliable digital auxiliary decision-making means for bridge closure construction is provided. BRIEF DESCRIPTION OF DRAWINGS

[0016] Hereinafter, the application will be described in more detail based on the embodiments and with reference to the accompanying drawings. In which: Figure 1 A flowchart of the application is shown; Figure 2 A bridge closure section point cloud data schematic diagram of the application is shown; Figure 3 A bridge closure opening point cloud data schematic diagram of the application is shown; Figure 4 A schematic diagram of the virtual pre-assembly of the application is shown; Figure 5 A bridge closure section connection end face region data segmentation process schematic diagram of the application is shown; Figure 6 A closure opening connection end face region data segmentation process schematic diagram of the application is shown; Figure 7 A point cloud data schematic diagram of the bridge closure section connection end face data after irrelevant point clouds are removed by the DBSCAN algorithm of the application is shown; Figure 8 A point cloud data schematic diagram of the bridge closure opening connection end face data after irrelevant point clouds are removed by the DBSCAN algorithm of the application is shown; Figure 9 A feature corner point result schematic diagram finally fitted by the application is shown; Figure 10 A feature corner point fitting mode of the application is shown.

[0017] In the drawings, the same components use the same reference numerals. The drawings are not in actual proportion. DETAILED DESCRIPTION

[0018] The application will be further described below with reference to the accompanying drawings.

[0019] The application provides a virtual pre-assembly method for a bridge closure section, as shown in the specification, comprising the following steps: Figure 1 Step S1: Obtain point cloud data of the closure section and the closure opening by using a land-based three-dimensional laser scanner, and pre-process the point cloud data to obtain pre-processed closure section point cloud data and closure opening point cloud data; Step S2: Extract connection cross-section region data of the closure section and the closure opening based on the dichotomy method, and determine the cutting result reservation mode according to the target region direction to obtain connection end face region data; Step S3: Further extract connection end face point cloud data from the connection end face region by combining neighborhood search and accurate connection end face data extraction of the projection plane; Step S4: Perform noise reduction processing on the extracted connection end face point cloud data of the closure section and the closure opening; Step S5: Extract the outer boundary of the connection end face and fit the feature corner points based on the point cloud boundary extraction algorithm, the clustering algorithm and the straight line fitting algorithm to obtain all key corner points on each connection end face, and finally obtain the feature corner point data set of the closure section and the closure opening; Step S6: Establish a feature point three-dimensional model based on the feature corner points of the closure section and the closure opening, and simulate the final state of hoisting and docking by using a virtual pre-assembly method; Step S7: Automatically calculate the closure section cutting amount based on the feature corner points and the connection end face normal vector of the closure opening; In step S1, the pre-processing includes registration, removal of environmental noise point cloud and unified coordinate processing. In one embodiment, as shown in the specification, Figure 5 In step S2, the dichotomy method is used to extract the connection cross-section region data of the closure section, which comprises the following steps: Step S211: Centralize the closure section point cloud coordinates, calculate the point cloud centroid, and take the centroid O as the coordinate origin; Step S212: Determine the initial cutting plane, perform principal component analysis (PCA) on the closure section point cloud data, then select a suitable principal plane and move it through the point cloud centroid to determine the initial cutting plane ; Step S213: Perform the first division, and divide the closure section point cloud into and two parts of point cloud based on the initial cutting plane ; Step S214: Perform the second division, first calculate and the new centroid coordinates of the two point clouds, and record them as and , and calculate the vectors and​ , respectively and PCA is performed on the two point clouds, respectively, and the third principal plane passing through the centroids of and is taken as the segmentation plane for the second segmentation of the point clouds and The second segmentation is performed, and four sub-point cloud data , , and are obtained, the order of which represents the relative positions between the sub-point clouds, and the centroid coordinates of the four sub-point cloud data are , , and , respectively , , and are calculated, and then the four vectors are respectively multiplied by the corresponding vectors and , and the point cloud data corresponding to the centroid with a positive calculation result is taken as the data to be reserved, i.e., the data to be reserved after the second segmentation is and ; Step S215, iteratively perform the second segmentation step until the set number of iterations 5 is reached to prevent over-segmentation; As shown in Figure 6 , the data extraction of the connection section of the joint is performed based on the bisection method, including: Step S221, center the point cloud coordinates of the joint, calculate the centroid of the point cloud, and take the centroid O as the coordinate origin; Step S222, determine the initial segmentation plane, perform principal component analysis on the point cloud data of the joint, then select a principal plane and translate it through the centroid of the point cloud, thereby determining the initial segmentation plane ; Step S223, perform the first segmentation, divide the point cloud of the joint into and two parts of point clouds according to the initial segmentation plane ; Step S224, perform the second segmentation, first calculate the new centroid coordinates of L1 and R1, respectively, and record them as and , and calculate the vectors and , respectively and perform principal component analysis on the two point clouds, respectively, and take the third principal plane passing through the centroids of and The third principal plane of the centroid is used as the segmentation surface for the second point cloud segmentation. and Then, a second segmentation is performed to obtain four sub-point cloud data. , , as well as The order represents the relative positions between the sub-point clouds, and the centroid coordinates of the four sub-point cloud data are respectively... , , as well as Calculate the centroid vector respectively , , as well as Then these four vectors are respectively compared with their corresponding vectors. and Perform a vector dot product, and keep the point cloud data corresponding to the centroids with positive results as the retained data. That is, the point cloud data retained after the second cut is... and ; Step S225: Iterate the second cutting step until the set number of iterations, 4, is reached to prevent excessive iteration and segmentation; In one embodiment, in step S3, the same processing method is applied to the obtained data of the connection end face region between the closure segment and the closure opening, and the connection end face point cloud data is extracted from the connection end face region data, including: Step S311: Extract the and Data, calculate the point cloud obtained after the (i-1)th cut. and The corresponding centroids and The centroids of the point clouds Li-1 and Ri-1 obtained by the i-th cut and The resulting vector and ; Step S312: Transformation from cutting plane to projection plane. The planes PLi and PRi used for the i-th iteration segmentation are respectively aligned along the vector... and The direction is translated by a factor of ten, thus constructing two projection planes. and ; Step S313: Project the target point cloud, projecting point clouds Li-1 and Ri-1 onto the projection plane respectively. and Above, the projected point cloud is obtained. and ; Step S314, extracting the nearest points on the source point cloud, respectively traversing the point data of and , and searching for the K nearest points to the projection points on the projection surface on the point cloud and using the K-d tree nearest neighbor algorithm; In one embodiment, as shown in Figure 7 and Figure 8 , the extracted joint segment and joint mouth connecting end surface point cloud data are denoised, and the steps are as follows: Step S411, repeating the point removal algorithm to process the extracted data; Step S412, using a clustering algorithm to cluster each connecting end surface data of the joint segment and the joint mouth, and then screening out the connecting end surface clustering data belonging to the joint segment; In one embodiment, as shown in Figure 9 and 10 , the steps of fitting the connecting end surface outer boundary extraction and feature corner point are as follows: Step S511, using a point cloud boundary extraction algorithm to extract the outer boundary point cloud of each connecting interface data of the joint segment and the joint mouth after denoising; Step S512, using a clustering algorithm to cluster and identify straight line data in the connecting interface data, using a straight line fitting algorithm to fit each straight line point cloud cluster, and setting a k value to filter point cloud clusters that do not meet the requirements; Step S513, initializing the start point and end point of each connecting interface outer boundary data fitted straight line; Step S514, rearranging the start point and end point of each straight line in clockwise order; Step S515, according to the obtained sorting result, calculating the nearest two points of each straight line and the next straight line, and then calculating the midpoint of the two points as the intersection point of the two straight lines, until six straight line intersection point three-dimensional coordinate data are obtained; In one embodiment, a feature point three-dimensional model is established, and the steps of simulating hoisting and docking using a virtual pre-assembly method are as follows: Step S611, respectively establishing the feature point model of the two connecting interfaces of the joint segment and the feature point model of the two connecting interfaces of the joint mouth; Step S612, based on the registration algorithm, keeping the joint mouth data position unchanged, and registering the joint segment feature point data into the joint mouth data, and the loss function is:

[0020] wherein n is the total number of joint segment feature corner points, With In each range of i = 1 to n, one-to-one correspondence, Indicates the Euclidean norm; In one embodiment, step S7 comprises: Calculate the projection length between each point pair along the corresponding closure connection surface normal vector direction as the cutting amount of the closure section; It should be noted that in the embodiment, as shown in Figure 4 As shown in the figure, taking a space arch tower cable-stayed bridge as an example, combined with virtual pre-assembly technology and high-precision feature point extraction algorithm, based on three-dimensional laser scanner to obtain the point cloud data of the closure section and the closure mouth in the built state, by introducing the connection interface data extraction process based on the dichotomy idea, combined with principal component analysis, nearest neighbor search and boundary feature point fitting technology, the automatic establishment of the spatial feature point model of the bridge closure section and the closure mouth is realized, and then through the virtual pre-assembly technology, the cutting amount at each feature point position is automatically calculated according to the feature point model registration alignment, that is, the problems of low efficiency, high error and high dependence on experience existing in the traditional total station + manual alignment mode are effectively solved, the precision and efficiency of the bridge closure section cutting construction are significantly improved, the precision and timeliness in the construction process are significantly improved, and a reliable digital auxiliary decision-making means for bridge closure construction is provided; In one embodiment, when collecting three-dimensional point cloud data by using multiple land three-dimensional laser scanners, the point cloud data in the connection end surface area is encrypted, the point cloud data obtained by adjacent scanning stations is partially overlapped, and the input data is as shown in Figure 2 And Figure 3 As shown in the figure; In one embodiment, in step S5, the boundary extraction algorithm includes but is not limited to the latitude and longitude line scanning method, the clustering algorithm includes but is not limited to the Gaussian mixture model clustering method, and the straight line fitting algorithm includes but is not limited to the RANSAC algorithm and the least square method.

[0021] In the description of the present application, it should be understood that the terms "upper", "lower", "bottom", "top", "front", "back", "inner", "outer", "left", "right" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore cannot be understood as a limitation on the present application.

[0022] While the application has been described with reference to particular embodiments thereof, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present application. It will be apparent to those skilled in the art that numerous modifications can be made within the scope of the present application as defined by the appended claims. It is intended that all such modification fall within the spirit and scope of the present application. It will be understood that the features described in connection with one embodiment can be used in connection with another embodiment.

Claims

1. A method for virtual pre-erection of a bridge closure segment, characterized in that, The method comprises the following steps: Obtaining three-dimensional point cloud data of the closure section and the closure gap, and preprocessing the obtained three-dimensional point cloud data; Based on the bisection method, the connection section area data of the closure section and the connection section area data of the closure gap are extracted, and the cutting result reservation mode is determined according to the target area direction to obtain the connection end surface area data; Combining field search with the accurate connection end surface data extraction of the projection surface, the connection end surface point cloud data is extracted from the connection end surface area; Based on the point cloud boundary extraction algorithm, the clustering algorithm and the straight line fitting algorithm, the connection end surface outer boundary extraction and the characteristic corner fitting are performed to obtain all the key corner points on each connection end surface, and finally the characteristic corner point data set of the closure section and the closure gap is obtained; The characteristic point three-dimensional model is constructed, and the virtual pre-assembly method is used to simulate the final state of hoisting and docking to calculate the closure section fitting amount.

2. The method for virtual pre-erection of a bridge closure segment according to claim 1, characterized in that, The three-dimensional point cloud data of the closure section and the closure gap is collected by using multiple land-based three-dimensional laser scanners, and the connection end surface area is collected with encrypted point data during the collection stage. The point cloud data obtained by adjacent scanners is overlapped.

3. The method for virtual pre-erection of a bridge closure segment according to claim 1, wherein, The obtained three-dimensional point cloud data is preprocessed, including: The three-dimensional point cloud data is registered, the environmental noise point cloud is removed, and the coordinates are unified.

4. The method for virtual pre-erection of a bridge closure segment according to claim 1, wherein, Based on the bisection method, the connection section area data of the closure section is extracted, including: The point cloud coordinate of the closure section is centralized, the point cloud centroid is calculated, and the centroid O is taken as the coordinate origin; Determine the initial cutting plane, perform principal component analysis on the point cloud data of the joint closure section, then select a principal plane and translate it to pass through the point cloud centroid, thereby determining the initial cutting plane ; performing a first segmentation to an initial cut surface For the division, the point cloud of the closure segment is divided into and two parts of point cloud; The second segmentation is performed, first calculating and the new centroid coordinates of the two point clouds, and recording as and , and calculating the vectors and , respectively performing principal component analysis on and the two point clouds, and taking the third principal plane of the centroids of and as the segmentation plane of the second point cloud segmentation and , and performing the second segmentation, obtaining four sub-point cloud data , , and , the order representing the relative positions between the sub-point clouds, the centroid coordinates of the four sub-point cloud data being , , and , respectively calculating the centroid vectors , , and , and then performing vector dot product of the four vectors with the corresponding vectors and , and taking the point cloud data corresponding to the centroid with a positive calculation result as the data to be retained; The second cutting step i is iterated twice.

5. The method for virtual pre-erection of a bridge closure segment according to claim 4, characterized in that, Based on the bisection method, the connection section area data of the closure gap is extracted, including: The point cloud coordinate of the closure gap is centralized, the point cloud centroid is calculated, and the centroid O is taken as the coordinate origin; determining an initial cutting plane, performing principal component analysis on the closure point cloud data, then selecting a principal plane and translating it to pass through the point cloud centroid, thereby determining an initial cutting plane ; performing a first segmentation to an initial cut surface For the division, the point cloud of the closure joint is divided into and two parts of point clouds; The second segmentation is performed, first, the new centroid coordinates of the two point clouds L1 and R1 are calculated respectively, and recorded as With , and the vector With , respectively, the principal component analysis is performed on And The third principal plane of the centroids of And is taken as the segmentation plane of the second point cloud segmentation And , the second segmentation is performed, obtaining four sub-point cloud data , , And , the order represents the relative position between the sub-point clouds, the centroid coordinates of the four sub-point cloud data are , , And , the centroid vectors , , And are calculated respectively, then the four vectors are respectively multiplied with the corresponding vectors And , and the point cloud data corresponding to the centroid with a positive calculation result is taken as the reserved data; The second cutting step i is iterated twice.

6. A method of virtual pre-erection of a bridge closure segment according to claim 4 or 5, characterized in that, The connection end surface point cloud data is extracted from the connection end surface area, including: In the remaining point cloud data, the point cloud obtained after the i-1th cutting is calculated and The corresponding centroid and The centroid of the point cloud Li-1 and Ri-1 obtained by the i-th cutting and The vector formed by and ; The conversion of the cutting planes to the projection planes, the planes PLi and PRi used in the i-th iteration of the segmentation are translated along the direction of the vectors and by a factor of ten times the vector length, so as to construct two projection planes and ; Target point cloud projection, projecting point clouds Li-1 and Ri-1 to the projection plane respectively and , to obtain projected point clouds and ; Extract the nearest point on the source point cloud, respectively traverse the point cloud and Point data, and use the nearest neighbor algorithm to search for the K nearest points on the point cloud and The projection point on the projection plane is closest.

7. The method for virtual pre-erection of a bridge closure segment according to claim 1, wherein, The extracted connection end surface point cloud data of the closure section and the closure gap is also subjected to noise reduction processing.

8. The method for virtual pre-erection of a bridge closure segment according to claim 7, characterized in that, The steps of the noise reduction processing are as follows: The extracted data is processed based on the distance threshold repeat point removal algorithm; The clustering algorithm is used to cluster each connection end surface data of the closure section and the closure gap, and then the connection end surface clustering data belonging to the closure section is screened out.

9. The method for virtual pre-erection of a bridge closure segment according to claim 7, wherein, The steps of extracting characteristic corner points from the closure section data and the closure gap data are as follows: The outer boundary point cloud of each connection interface data of the closure section and the closure gap after noise reduction processing is extracted using the point cloud boundary extraction algorithm; The straight line data in the connection interface data is identified using the clustering algorithm, the straight line fitting algorithm is used to fit each straight line point cloud cluster, and a k value is set to filter the point cloud clusters that do not meet the requirements; The start point and the end point of each connection interface outer boundary data fitting straight line are initialized; The start point and the end point of each straight line are rearranged in order; According to the obtained sorting result, the nearest two points of each straight line to the next straight line are calculated, and then the midpoint of the two points is calculated as the intersection point of the two straight lines, until all the straight line intersection point three-dimensional coordinate data is obtained.

10. The method for virtual pre-erection of a bridge closure segment according to claim 1, wherein, The steps of establishing the characteristic point three-dimensional model and using the virtual pre-assembly method to simulate the hoisting and docking are as follows: The characteristic point model of the two connecting interfaces of the closure section and the characteristic point model of the two connecting interfaces of the closure gap are respectively established; Based on the registration algorithm, the closure section characteristic point data is registered into the closure gap data while keeping the closure gap data position unchanged, and the loss function is: wherein n is the total number of characteristic corner points of the closure section, with corresponding to each i = 1 to n, denotes the Euclidean norm; The projection length of each point pair along the corresponding closure gap connecting surface normal vector direction is calculated and taken as the cutting amount of the closure section.

Images