A welding plate point cloud hole repairing method based on point cloud rasterization

Abstract

The application discloses a welding plate point cloud hole repairing method based on point cloud rasterization, and first acquires a welding plate point cloud image to be repaired, creates a point cloud bounding box, divides the point cloud bounding box into a plurality of grids with the same size and point cloud density, fits a welding plate plane equation, and corresponds the grids to the welding plate according to the grid numbers, traverses the point cloud, divides points into corresponding grids, traverses the grids, does not need to repair points when the grid is not empty, repairs points in the grid when the grid is empty, and thus the hole repairing is realized. In the process of repairing the welding plate point cloud, the structural features of the welding plate point cloud before repairing are fully considered, the repairing process is ensured to have no influence on the original welding plate point cloud structure, and meanwhile, the process is simple and convenient; the overall density information of the welding plate point cloud is added, so that the overall density of the welding plate point cloud before and after repairing is basically unchanged, and the shape information of the welding plate plane is added, so that the repairing process is simple and fast.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of welding plate point cloud hole repair, and particularly relates to a welding plate point cloud hole repair method based on point cloud rasterization. BACKGROUND

[0002] Point cloud is a collection of massive points describing the surface characteristics of an object, which is usually obtained by using a laser radar or a scanner. When a contour scanner is used to obtain a welding plate point cloud, due to the accuracy of the visual sensor, or due to the rust on the surface of the welding plate and the reflection of the metal material, and the problems such as possible occlusion, the position of part of the points in the final scanning result is offset, forming a hole, and the hole will adversely affect the subsequent feature recognition and three-dimensional reconstruction process, greatly affecting the effect of the weld reconstruction, and causing difficulty in the accuracy of the subsequent weld feature extraction.

[0003] Therefore, how to repair the welding plate point cloud with a hole defect is a problem to be solved by those skilled in the art. SUMMARY

[0004] The present application is proposed to solve the problems of the prior art, and provides a welding plate point cloud hole repair method based on point cloud rasterization, which fully considers the structural characteristics of the original point cloud, ensures that the repair process does not affect the point cloud structure, and is simple and fast.

[0005] The technical scheme for achieving the purpose of the present application is as follows:

[0006] A welding plate point cloud hole repair method based on point cloud rasterization comprises the following steps:

[0007] Step S1, obtaining an image of the welding plate point cloud to be repaired;

[0008] Step S2, creating a point cloud bounding box to identify the boundary of the welding plate point cloud, and using a two-dimensional bounding box to completely surround the welding plate point cloud;

[0009] Step S3, calculating the point cloud density, dividing the two-dimensional bounding box into a plurality of grids, and the size of the grid is the same as the density of the welding plate point cloud;

[0010] Step S4, fitting a welding plate plane equation according to the welding plate point cloud, and corresponding the grid to the welding plate according to the grid number;

[0011] Step S5, traversing the points in the welding plate point cloud, and dividing the points into corresponding grids according to the coordinates of the points;

[0012] Step S6, traversing the grids, when the grid is not empty, no point needs to be filled, and when the grid is empty, a point is filled in the grid.

[0013] Further, in the step S1, the profile scanner is installed on the moving platform, and the moving platform is controlled to move along the weld seam at a constant speed, so as to drive the sensor of the profile scanner to scan and acquire data, and convert the data into the image of the weld plate point cloud.

[0014] Further, the actual distance between two points in the weld plate point cloud is calculated by substituting the length l of the laser of the profile scanner, the number n of data columns, the sampling frequency f of the profile scanner, and the moving speed v of the moving platform into the following formula, so as to convert the data acquired by the sensor of the profile scanner into the image of the weld plate point cloud,

[0015]

[0016] y_step=v·f

[0017] wherein x_step and y_step represent the actual distance between two points in the x and y coordinates.

[0018] Further, in the step S2, the four corner points of the two-dimensional bounding box are determined by calculating the maximum value in the x and y directions of the entire weld plate point cloud, so as to create the point cloud bounding box.

[0019] Further, in the step S3, the average distance of each point to its neighbor points is calculated by using the K-Nearest Neighbor method and the following formula, so as to estimate the point cloud density,

[0020]

[0021] wherein, represents the average density of the point cloud, d p represents the minimum distance of the point p to other points.

[0022] Further, in the step S4, the least square method is used to fit the weld plate plane equation, and the fitting result includes four coefficients a, b, c, and d describing the plane, the form of the weld plate plane equation is ax+by+cz+d=0, the position of the split line of the two weld plates is calculated by the weld plate plane equation, and the weld plate corresponding to the grid is determined according to the position of the split line and the number of the grid.

[0023] Further, in the step S5, according to the coordinates of the point and the number of the grid, the point is divided into the grid by using the following formula,

[0024]

[0025]

[0026] wherein x_grid and y_grid represent the x and y numbers of the grid to which the point belongs, x miny min x, y minimum coordinate values of valid points representing the whole welding plate point cloud, size is the size of the grid.

[0027] Further, in the step S6, the coordinates x, y, z of the points to be filled are calculated according to the following formula,

[0028] x = (x_grid + 0.5) * size + x min

[0029] y = (y_grid + 0.5) * size + y min

[0030]

[0031] With the above technical solutions, the present application has the following beneficial effects:

[0032] (1) The present application fully considers the structural characteristics of the welding plate point cloud before repair in the process of repairing the welding plate point cloud, ensures that the repair process will not affect the original welding plate point cloud structure, and is simple and convenient; by adding the overall density information of the welding plate point cloud, the overall density of the welding plate point cloud before and after repair is basically unchanged, and by adding the shape information of the welding plate plane, the repair process is simple and fast.

[0033] (2) The present application ensures that the obtained data is more accurate by moving the platform to drive the contour scanner to scan at a constant speed.

[0034] (3) The cavity repair of the present application combines multiple algorithms, such as K-nearest neighbor method and least square method, so that the processing of related data is more objective, and ensures that the welding plate point cloud after cavity repair is more in line with the actual situation, and is more conducive to subsequent feature recognition and weld reconstruction. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to make the content of the present application more easily understood, the following will further explain the present application in detail according to specific embodiments and in combination with the accompanying drawings, in which:

[0036] Figure 1 is a flow chart of the present application. DETAILED DESCRIPTION

[0037] In order to better understand the above technical solutions, the above technical solutions will be explained in detail in combination with the drawings and specific embodiments of the specification. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0038] (Example 1)

[0039] As Figure 1 shown in the welding plate point cloud hole repairing method based on point cloud rasterization, the structure characteristics of the welding plate point cloud before repairing are fully considered in the process of repairing the welding plate point cloud, so as to ensure that the repairing process will not affect the structure of the original welding plate point cloud, and the method is simple and convenient, which specifically includes the following steps:

[0040] Step S1, obtaining an image of the welding plate point cloud to be repaired;

[0041] First, install the contour scanner on the mobile platform, control the mobile platform to move at a predetermined speed along the weld seam at a uniform speed, thereby driving the sensor of the contour scanner to scan and obtain data, and use the mobile platform to ensure that the obtained data is more accurate. Since the data type read directly from the sensor is not a point cloud data type, it is necessary to first convert the format of the data, that is, to calibrate the actual distance between two points in the welding plate point cloud x, y coordinates, by substituting the laser length l of the contour scanner, the data column number n, the sampling frequency f of the contour scanner and the movement speed v of the mobile platform into the following formula,

[0042]

[0043] y_step=v·f

[0044] Wherein, x_step, y_step represents the actual distance between two points x, y coordinates, so as to convert the data obtained by the sensor scanning of the contour scanner into the image of the welding plate point cloud.

[0045] Step S2, creating a point cloud bounding box to identify the boundary of the welding plate point cloud, using a two-dimensional bounding box to completely surround the welding plate point cloud; by calculating the maximum and minimum values of the welding plate point cloud in x and y directions, the four corner points of the two-dimensional bounding box are determined, and the four maximum values are x_min, x_max, y_min, y_max, and the four corner points are [x_min, y_min], [x_min, y_max], [x_max, y_min], [x_max, y_max], so as to create a point cloud bounding box.

[0046] Step S3, using K nearest neighbor method, calculating the average distance of each point to its nearest neighbor according to the following formula, so as to estimate the point cloud density,

[0047]

[0048] Wherein, represents the average density of the point cloud, d p represents the minimum distance from point p to other points,

[0049] d pmin(dis(p,q)), q = 1, 2,..., n

[0050] where q represents other points, and dis(p, q) represents the distance from point p to other points.

[0051] The two-dimensional bounding box is divided into a plurality of grids according to the point cloud density, and the size of the grid is required to be the same as the density of the welding plate point cloud, so as to ensure that the point cloud structure is consistent before and after repair.

[0052] In step S4, a least square method is used to fit a welding plate plane equation according to the welding plate point cloud, and the fitting result includes four coefficients a, b, c, and d describing the plane. The form of the welding plate plane equation is ax+by+cz+d=0. The position of the split line of the two welding plates is calculated through the welding plate plane equation, and the welding plate corresponding to the grid is determined according to the position of the split line and the number of the grid, so as to correspond the grid to the welding plate.

[0053] In step S5, the points in the welding plate point cloud are traversed, and the points are divided into corresponding grids according to the coordinates of the points. The coordinates of the points and the number of the grids are calculated according to the following formula:

[0054]

[0055]

[0056] where x_grid and y_grid represent the x and y numbers of the grid to which the point belongs, x min ,y min represent the minimum x and y coordinate values of the effective points of the entire welding plate point cloud, and size is the size of the grid, so as to divide the points into corresponding grids.

[0057] In step S6, the grids are traversed. When the grid is not empty, no points need to be filled. When the grid is empty, points are filled in the grid. The coordinates x, y, and z of the points that need to be filled are calculated according to the following formula:

[0058] x = (x_grid + 0.5) * size + x min

[0059] y = (y_grid + 0.5) * size + y min

[0060]

[0061] The position of the point that needs to be filled is determined according to the calculation result, so as to complete the hole repair.

[0062] The application fully considers the structural features of the welding plate point cloud before repair, ensures that the repair process will not affect the original welding plate point cloud structure, and is simple and convenient; by adding the overall density information of the welding plate point cloud, the overall density of the welding plate point cloud before and after repair is basically unchanged, by adding the shape information of the welding plate plane, the repair process is simple and fast. In the process of hole repair, a variety of algorithms such as K nearest neighbor method and least square method are combined, so that the processing of related data is more objective, and the welding plate point cloud after hole repair is more consistent with the actual situation, and is more conducive to subsequent feature recognition and weld reconstruction.

[0063] The above specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the application. It should be understood that the above description is only a specific embodiment of the application and is not intended to limit the application. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the application should be included in the protection scope of the application.

Claims

1. A welding plate point cloud hole repairing method based on point cloud rasterization, characterized in that, The method comprises the following steps: In step S1, an image of a welding plate point cloud to be repaired is acquired, and the actual distance between two points in the welding plate point cloud is calibrated by substituting the laser length l of a profile scanner, the number of data columns n, the sampling frequency f of the profile scanner and the motion speed v of a moving platform into the following formula, so as to convert the data scanned by the sensor of the profile scanner into the image of the welding plate point cloud, y_step=v·f Wherein, x_step and y_step represent the actual distance between two points in x and y coordinates. In step S2, a point cloud bounding box is created, and the boundary of the welding plate point cloud is identified, and the welding plate point cloud is completely surrounded by a two-dimensional bounding box. In step S3, the point cloud density is calculated, and the average distance of each point to its neighbor point is calculated by using the following formula through the K nearest neighbor method, so as to estimate the point cloud density, wherein, represents the average density of the point cloud, d p represents the minimum distance of point p to other points, The two-dimensional bounding box is divided into a plurality of grids, and the size of the grid is the same as the density of the welding plate point cloud. In step S4, the welding plate plane equation is fitted according to the welding plate point cloud, and the grid is matched with the welding plate according to the number of the grid. In step S5, the points in the welding plate point cloud are traversed, and the points are divided into the corresponding grid according to the coordinates of the points. In step S6, the grid is traversed, and when the grid is not empty, no point needs to be supplemented, and when the grid is empty, a point is supplemented in the grid.

2. The welding plate point cloud hole repairing method based on point cloud rasterization according to claim 1, characterized in that: In step S1, the profile scanner is installed on the moving platform, and the moving platform is controlled to move uniformly along the weld, so as to drive the sensor of the profile scanner to scan and acquire data, and convert the data into the image of the welding plate point cloud.

3. The welding plate point cloud hole repairing method based on point cloud rasterization according to claim 1, characterized in that: In step S2, the four corner points of the two-dimensional bounding box are determined by calculating the maximum value of the x and y directions of the entire welding plate point cloud, so as to create the point cloud bounding box.

4. The welding plate point cloud hole repairing method based on point cloud rasterization according to claim 1, characterized in that: In step S4, the least square method is used to fit the welding plate plane equation, and the fitting result includes four coefficients a, b, c and d describing the plane, and the form of the welding plate plane equation is ax+by+cz+d=0. The position of the split line between two welding plates is calculated by the welding plate plane equation, and the welding plate corresponding to the grid is determined according to the position of the split line and the number of the grid.

5. The welding plate point cloud hole repairing method based on point cloud rasterization according to claim 1, characterized in that: In step S5, the points are divided into the grid according to the following formula according to the coordinates of the points and the number of the grid, where x_grid and y_grid represent the x and y numbers of the grid that the point belongs to, x min ,y min represents the x, y minimum coordinate values of the valid points of the entire welding plate point cloud, and size is the size of the grid.

6. The welding plate point cloud hole repairing method based on point cloud rasterization according to claim 1, characterized in that: In step S6, the coordinates x, y and z of the point to be supplemented are calculated according to the following formula, x = (x_grid + 0.5) * size + x min y = (y_grid + 0.5) * size + y min

Images