Ultrasonic scanning fixture adjustment method based on point cloud data

Abstract

The present invention provides an ultrasonic scanning fixture adjustment method based on point cloud data, which belongs to the field of image processing technology, and specifically includes: using target coordinate data, using an overdetermined method to solve to obtain multiple groups of plane fitting results of the carrier surface, based on the distance to the carrier surface, iterative processing is performed to determine the target coordinate data with the largest number of retained internal points, using the target coordinate data with the largest number of retained internal points to refit the equation for the final surface fitting result, based on the equation for the final surface fitting result, calculating the posture information of the normal vector of the actual carrier surface and the normal vector of the ideal plane, based on the posture information, obtaining the roll angle around the scanning axis and the pitch angle around the stepping axis, and determining the screw adjustment amount and rotation angle of the ultrasonic scanning fixture based on the roll angle and the pitch angle, thereby improving the efficiency of the fixture adjustment processing.

Description

Technical Field

[0001] The present invention belongs to the technical field of image processing, and in particular relates to an ultrasonic scanning fixture adjustment method based on point cloud data. Background Art

[0002] Ultrasonic scanning microscopy equipment uses ultrasonic waves to image the product under test. Scanning imaging methods primarily include time-of-flight imaging and amplitude imaging. The scanning imaging system consists of a water tank, within which a fixture is placed. The fixture supports the product under test. If the fixture is uneven, the propagation path length of the ultrasonic wave will vary, affecting the accuracy of the time-of-flight measurement and, consequently, the image quality. Therefore, ultrasonic scanning imaging equipment requires fixture leveling.

[0003] Existing technical solutions often use images for adjustment processing, through single-point sampling and then a simple size comparison of the sampled data. This is too random and has insufficient noise resistance. Secondly, there is a coupling relationship between the two inclination angles of the plane. During the adjustment process, after the inclination angle in one direction meets the index, the angle in the other direction exceeds the tolerance, and multiple cycles of adjustment are required.

[0004] In order to solve the above technical problems, the present application provides an ultrasonic scanning fixture adjustment method based on point cloud data. Summary of the Invention

[0005] To achieve the purpose of the present invention, the present invention adopts the following technical solutions:

[0006] Specifically, the present application provides an ultrasonic scanning fixture adjustment method based on point cloud data, which specifically includes:

[0007] Divide the vehicle surface to obtain a coordinate point set consisting of N three-dimensional point cloud coordinates, perform centroid normalization on the coordinate point set, and obtain target coordinate data;

[0008] Using the target coordinate data, an overdetermined method is used to solve multiple sets of plane fitting results for the vehicle surface, and based on the distance from the vehicle surface, an iterative process is performed to determine the target coordinate data with the largest number of retained inliers, and the target coordinate data with the largest number of retained inliers is used to refit the equation for obtaining the final surface fitting result;

[0009] Based on the equation of the final surface fitting result, the attitude information of the normal vector of the actual vehicle surface and the normal vector of the ideal plane is calculated, and the roll angle around the scanning axis and the pitch angle around the stepping axis are obtained based on the attitude information;

[0010] Based on the roll angle and the pitch angle, the screw adjustment amount and the rotation angle of the ultrasonic scanning fixture are determined.

[0011] The beneficial effects of the present invention are:

[0012] The target coordinate data with the largest number of retained inliers is determined through iterative processing. The target coordinate data with the largest number of retained inliers is used to refit the equation for the final surface fitting result, thereby reducing the interference of noise data, suppressing the observation noise of point cloud data, and eliminating gross error point cloud data, making the observation results more accurate.

[0013] Based on the roll angle and pitch angle, the screw adjustment amount and rotation angle of the ultrasonic scanning fixture are determined, and automatic identification and processing of the screw adjustment amount and rotation angle of the ultrasonic scanning fixture based on point cloud data are realized. By calculating the posture information of the normal vector of the actual carrier surface and the normal vector of the ideal plane, the technical problem of repeated adjustment of single-point adjustment is avoided, thereby improving the adjustment processing rate of the screw.

[0014] A further technical solution is to divide the vehicle surface into sections, specifically including:

[0015] The position of the scanning axis and the position of the stepping axis are divided into a grid form according to a fixed interval, for example, 0.05 mm, that is, a uniform division method, to perform division processing on the carrier surface.

[0016] In another embodiment, the carrier surface is divided according to a double-Gaussian bell-shaped division in both the scanning axis position and the stepping axis position directions.

[0017] A further technical solution is that the three-dimensional point cloud coordinates are is the position of the scanning axis, is the position of the stepper axis, It is the altitude value calculated based on the flight time.

[0018] A further technical solution is that the normal vector of the actual vehicle surface is obtained by solving an equation of the final surface fitting result.

[0019] A further technical solution is that the roll angle around the scanning axis and the pitch angle around the stepping axis are determined in the form of Euler angles.

[0020] A further technical solution is that the calculation formula of the roll angle around the scanning axis is: in is the roll angle around the scan axis, is the y-axis normal vector of the actual vehicle plane, is the z-axis normal vector of the actual vehicle plane.

[0021] Furthermore, the screw adjustment amount of the ultrasonic scanning fixture includes adjustment amounts on the x-axis and the y-axis.

[0022] Specifically, the calculation formula for the adjustment amount of the x-axis is: , For the vehicle Length in the axial direction.

[0023] Furthermore, the calculation formula for the screw rotation angle is: Where D is the lead data of the screw, The adjustment amount of the x-axis or y-axis is determined by calculating the screw rotation angle of the x-axis or y-axis.

[0024] Furthermore, the rotation angle is pushed through the device interface and the operator manually completes the screw rotation adjustment, or is adjusted by an automatically operated electrical mechanical mechanism.

[0025] Other features and advantages will be described in the following description. The objectives and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description and drawings.

[0026] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the accompanying drawings.

[0028] Figure 1 It is a flow chart of an ultrasonic scanning fixture adjustment method based on point cloud data;

[0029] Figure 2 is a schematic diagram of ultrasonic testing processing;

[0030] Figure 3 is a flow chart of the equation for refitting to obtain the final plane fitting result. DETAILED DESCRIPTION

[0031] To help those skilled in the art better understand the technical solutions in this specification, the following will provide a clear and complete description of the technical solutions in the embodiments of this specification, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this specification without creative work should fall within the scope of protection of this specification.

[0032] Example 1

[0033] like Figure 1 As shown, the present application provides an ultrasonic scanning fixture adjustment method based on point cloud data, specifically comprising:

[0034] Divide the vehicle surface to obtain a coordinate point set consisting of N three-dimensional point cloud coordinates, perform centroid normalization on the coordinate point set, and obtain target coordinate data;

[0035] Specifically, such as Figure 2 As shown in FIG, a schematic diagram of the ultrasonic detection process of the present application includes (1) point cloud data acquisition:

[0036] Divide the vehicle surface into A set of coordinate points consisting of three-dimensional coordinates ,in: is the position of the scanning axis, is the position of the stepper axis, It is the altitude value calculated based on the flight time.

[0037] As an example, the vehicle surface is divided into and The two directions are divided into a grid form with a fixed spacing of, for example, 0.05 mm, i.e., a uniform division method. Other division methods with prior knowledge may also be used, such as a double Gaussian bell-shaped division, i.e., the closer to the center area, the greater the division density, and these division methods also fall within the scope of protection of the present invention.

[0038] (2) Point cloud data normalization:

[0039] Differences in equipment dimensions will lead to large calculation errors. and The value ranges from 0 to 450 mm. The numerical value of micrometer level will be and When performing calculations together, large numbers may swallow up small numbers. Using the centroid-based normalization method, the coordinate origin is moved to the geometric center of the point cloud to facilitate subsequent processing and calculation:

[0040] in, is the corresponding mean.

[0041] Using the target coordinate data, an overdetermined method is used to solve multiple sets of plane fitting results for the vehicle surface, and based on the distance from the vehicle surface, an iterative process is performed to determine the target coordinate data with the largest number of retained inliers, and the target coordinate data with the largest number of retained inliers is used to refit the equation for obtaining the final surface fitting result;

[0042] Include the following in the above steps:

[0043] (3) Carrier plane fitting:

[0044] For the carrier plane , can be obtained through the equation It can be further expressed as equation The above-mentioned carrier plane The equation can be solved accurately in an overdetermined manner using three sets of point cloud coordinates, or it can be solved in an overdetermined manner using multiple sets of point cloud coordinates through optimization methods. The latter is generally based on minimizing the square of the residual to obtain the equation coefficients .

[0045] The specific steps for solving the problem using the overdetermined method are:

[0046] Given a set of 3D point cloud data , where n is the number of point clouds, generally n>>3. For solving such overdetermined problems, the least squares method can be used to minimize the residual. The specific method is as follows:

[0047] 1. Plane equation , can be rewritten as z = ax + by + c;

[0048] 2. Based on n observation data point clouds, establish an error function ;

[0049] 3. Take the partial derivative of e with respect to a, b, and c and set the derivative to zero to solve the equation with coefficients {a, b, c}.

[0050] (4) Point cloud data noise suppression: There is noise interference in the point cloud acquisition process. Severe interference leads to outliers, also known as external points. In order to suppress the observation noise of point cloud data, the iterative weighted least squares method is further used to eliminate gross point cloud data and filter out the internal point data.

[0051] The specific method is as follows: 1. Randomly select three sets of point cloud coordinates , , , that is, uniquely determine a plane ; 2. Traverse all point cloud coordinates and plane Point clouds with distances less than the threshold are recorded as inliers; 3. Perform multiple iterations, for example 1000 times, and retain the plane with the largest number of inliers; 4. Use all the retained inliers and use the method in step (3) above to refit the final plane. Equation coefficient.

[0052] Based on the equation of the final surface fitting result, the attitude information of the normal vector of the actual vehicle surface and the normal vector of the ideal plane is calculated, and the roll angle around the scanning axis and the pitch angle around the stepping axis are obtained based on the attitude information;

[0053] Specifically, the above steps include the following:

[0054] (5) Vehicle posture calculation

[0055] According to the final plane The equation can be used to obtain the normal vector of the actual vehicle plane , the normal vector of the ideal plane is , from which the normal vector of the actual vehicle plane can be calculated Normal vector to the ideal plane The attitude information of . In the form of Euler angles, we can get the attitude information around the scanning axis. Roll angle , and around the stepper axis Pitch angle .

[0056] Based on the roll angle and the pitch angle, the screw adjustment amount and the rotation angle of the ultrasonic scanning fixture are determined.

[0057] The above steps specifically include the following:

[0058] (6) Calculation of screw adjustment amount.

[0059] Based on the aforementioned Euler angle form of attitude information, combined with the vehicle's physical size data, we can calculate Direction adjustment amount and Direction adjustment amount ,in: For the vehicle The length of the direction.

[0060] (7) Calculation of screw rotation angle.

[0061] Based on the adjustment amount calculated above, combined with the screw lead data , the screw rotation angle can be calculated .

[0062] Furthermore, the vehicle surface is divided into the following categories:

[0063] The position of the scanning axis and the position of the stepping axis are divided into a grid form according to a fixed interval, for example, 0.05 mm, that is, a uniform division method, to perform division processing on the carrier surface.

[0064] In another embodiment, the carrier surface is divided according to a double-Gaussian bell-shaped division in both the scanning axis position and the stepping axis position directions.

[0065] Specifically, the three-dimensional point cloud coordinates are is the position of the scanning axis, is the position of the stepper axis, It is the altitude value calculated based on the flight time.

[0066] It should be noted that the value range of N is more than 3, and is specifically determined according to the observation results of the historical observation noise of the point cloud data. The greater the number of historical observation noises, the larger the value of N.

[0067] Further, such as Figure 3 As shown, the equation for refitting to obtain the final plane fitting result includes:

[0068] Three groups of 3D point cloud coordinates are randomly selected and solved to uniquely determine the fitting result of a vehicle surface; all 3D point cloud coordinates are traversed and the 3D point cloud coordinates whose distance to the vehicle surface is less than a threshold are marked as inliers; multiple iterations are performed to determine the 3D point cloud coordinates of the vehicle surface that retains the largest number of inliers; the fitting result of the vehicle surface that retains the largest number of inliers is used to solve in an overdetermined manner and refit the equation coefficients of the final surface fitting result.

[0069] In another possible embodiment, during the iterative analysis process of the three-dimensional point cloud coordinates, obtaining the interior points in the three-dimensional point cloud coordinates, and determining the three-dimensional point cloud coordinates of the coefficients of the equation for fitting the final surface fitting result based on the number of interior points, specifically includes:

[0070] Taking the three-dimensional point cloud coordinates of the vehicle surface with the largest number of inliers as the target point cloud coordinates, determining the deviation of the number of inliers between other point cloud coordinates and the target point cloud coordinates, determining the reference point cloud coordinates among the other point cloud coordinates based on the deviation, obtaining the number of the reference point cloud coordinates, and determining the coordinate credible value of the target point cloud coordinates based on the deviation between the reference point cloud coordinates and the target point cloud coordinates;

[0071] It should be noted that when there are no reference point cloud coordinates for the target point cloud coordinates, a variety of division methods are used, including double Gaussian bell-shaped division and uniform division, and the acquisition of point cloud coordinates is continued. The point cloud coordinates with the largest coordinate credibility value in different division results are used as the three-dimensional point cloud coordinates for fitting the equation coefficients of the final surface fitting result.

[0072] It should also be noted that if the number of reference point cloud coordinates does not meet the requirements, that is, it is greater than the threshold, it means that the target point cloud coordinates are likely to be interference point clouds. Therefore, on this basis, a variety of division methods are adopted, including double Gaussian bell-shaped division and uniform division, and the point cloud coordinates are continued to be acquired. The point cloud coordinates with the largest coordinate credibility value in different division results are used as the three-dimensional point cloud coordinates for fitting the equation coefficients of the final surface fitting result.

[0073] Even if the number of reference point cloud coordinates meets the requirements, it is necessary to further determine whether the coordinate credibility value of the target point cloud coordinates meets the requirements. When the coordinate credibility value of other target point cloud coordinates is less than the preset coordinate credibility threshold, a variety of division methods are adopted, including double Gaussian bell-shaped division and uniform division. The acquisition of point cloud coordinates continues, and the point cloud coordinates with the largest coordinate credibility value in different division results are used as the three-dimensional point cloud coordinates for fitting the equation coefficients of the final surface fitting result. In other cases, proceed to the next step.

[0074] Taking the reference point cloud coordinates as the target point cloud coordinates, and determining the coordinate credible value of the reference point cloud coordinates based on the number of reference point cloud coordinates when the reference point cloud coordinates are taken as the target point cloud coordinates and the deviation between the reference point cloud coordinates and the target point cloud coordinates;

[0075] It can be understood that when the sum of the coordinate credibility values ​​of the reference point cloud coordinates is less than the preset threshold, the distribution of the target point cloud coordinates is relatively discrete, and the deviation in the number of internal points between the target point cloud coordinates and other point cloud coordinates is large. Therefore, it means that the target point cloud coordinates at this time are likely to be interference point clouds. Therefore, on this basis, a variety of division methods are adopted, including double Gaussian bell-shaped division and uniform division, and the acquisition of point cloud coordinates is continued. The point cloud coordinates with the largest coordinate credibility value in different division results are used as the three-dimensional point cloud coordinates for fitting the equation coefficients of the final surface fitting result.

[0076] If the sum of the coordinate credible values ​​of the reference point cloud coordinates is not less than the preset threshold, it is also necessary to determine whether the number of reference point cloud coordinates whose coordinate credible values ​​meet the requirements meets the requirements, that is, whether the number of reference point cloud coordinates whose coordinate credible values ​​are greater than the preset coordinate credible threshold is greater than the preset number threshold. If the number of reference point cloud coordinates whose coordinate credible values ​​are greater than the preset coordinate credible threshold is not greater than the preset number threshold, then the distribution of the target point cloud coordinates is relatively discrete, and the deviation in the number of inner points between the coordinates of other point cloud coordinates is large. Therefore, it means that the probability that the target point cloud coordinates at this time are interference point clouds is relatively high. Therefore, on this basis, a variety of division methods are adopted, including double Gaussian bell-shaped division and uniform division, and the point cloud coordinates are continued to be acquired, and the point cloud coordinates with the largest coordinate credible value in different division results are used as the three-dimensional point cloud coordinates for fitting the equation coefficients of the final surface fitting result.

[0077] Based on the coordinate credible value of the target point cloud coordinates and the coordinate credible value of the reference point cloud coordinates, the three-dimensional point cloud coordinates of the equation coefficients for fitting the final surface fitting result are determined.

[0078] It can be understood that the reference point cloud coordinates are other point cloud coordinates whose deviation in the number of inliers from the target point cloud coordinates is less than a preset deviation threshold.

[0079] Specifically, the method for determining the coordinate credible value is:

[0080] Determining the deviation rates of different reference point cloud coordinates based on the ratio of the deviation between the reference point cloud coordinates and the target point cloud coordinates to the number of inliers of the target point cloud coordinates, and determining the credible values ​​of different reference point cloud coordinates based on the difference between 1 and the deviation rate;

[0081] The coordinate confidence value is determined based on the sum of confidence values ​​of different reference point cloud coordinates.

[0082] It can be understood that, based on the coordinate credible value of the target point cloud coordinates and the coordinate credible value of the reference point cloud coordinates, determining the three-dimensional point cloud coordinates of the equation coefficients for fitting the final surface fitting result specifically includes:

[0083] When the sum of the coordinate credibility values ​​of the reference point cloud coordinates is less than the preset threshold, the distribution of the target point cloud coordinates is relatively discrete, and the deviation in the number of internal points between the target point cloud coordinates and other point cloud coordinates is large. Therefore, it is explained that the probability that the target point cloud coordinates at this time are interference point clouds is relatively high. Therefore, on this basis, a variety of division methods are adopted, including double Gaussian bell-shaped division and uniform division, to continue to obtain the point cloud coordinates, and use the point cloud coordinates with the largest coordinate credibility value in different division results as the three-dimensional point cloud coordinates for fitting the equation coefficients of the final surface fitting result.

[0084] When the sum of the coordinate credible values ​​of the reference point cloud coordinates is not less than a preset threshold, and when the coordinate credible value of the target point cloud coordinates is greater than the preset credible threshold, the target point cloud coordinates are used as the three-dimensional point cloud coordinates of the equation coefficients for fitting the final surface fitting result;

[0085] When the coordinate credible value of the target point cloud coordinates is not greater than a preset credible threshold, obtaining a difference between the coordinate credible value of the reference point cloud coordinates and the coordinate credible value of the target point cloud coordinates, and using it as a credible deviation value;

[0086] When there are reference point cloud coordinates with a credible deviation value greater than a preset credible deviation threshold, the reference point cloud coordinates with the largest credible deviation value are used as the three-dimensional point cloud coordinates of the equation coefficients for fitting the final surface fitting result;

[0087] When there are no reference point cloud coordinates with a credible deviation value greater than the preset credible deviation threshold, multiple divisions are used to continue acquiring the point cloud coordinates, and the point cloud coordinates with the largest coordinate credible value in different division results are used as the three-dimensional point cloud coordinates for fitting the equation coefficients of the final surface fitting result.

[0088] Furthermore, the normal vector of the actual vehicle surface is obtained by solving the equation of the final surface fitting result.

[0089] Specifically, the roll angle around the scan axis and the pitch angle around the step axis are determined in the form of Euler angles.

[0090] Furthermore, the calculation formula of the roll angle around the scanning axis is: in is the roll angle around the scan axis, is the y-axis normal vector of the actual vehicle plane, is the z-axis normal vector of the actual vehicle plane.

[0091] Furthermore, the screw adjustment amount of the ultrasonic scanning fixture includes adjustment amounts on the x-axis and the y-axis.

[0092] Specifically, the calculation formula for the adjustment amount of the x-axis is: , For the vehicle Length in the axial direction.

[0093] Furthermore, the calculation formula for the screw rotation angle is: Where D is the lead data of the screw, The adjustment amount of the x-axis or y-axis is determined by calculating the screw rotation angle of the x-axis or y-axis.

[0094] Furthermore, the rotation angle is pushed through the device interface and the operator manually completes the screw rotation adjustment, or is adjusted by an automatically operated electrical mechanical mechanism.

[0095] The various embodiments in this specification are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the device, apparatus, and non-volatile computer storage medium embodiments are generally similar to the method embodiments, so their descriptions are relatively simplified. For relevant details, refer to the descriptions of the method embodiments.

[0096] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0097] The foregoing description is merely one or more embodiments of this specification and is not intended to limit this specification. It will be apparent to those skilled in the art that various modifications and variations may be made to one or more embodiments of this specification. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of one or more embodiments of this specification are intended to be within the scope of the claims of this specification.

Claims

1. A method for adjusting an ultrasonic scanning fixture based on point cloud data, characterized in that: Specifically include: Divide the vehicle surface to obtain a coordinate point set consisting of N three-dimensional point cloud coordinates, perform centroid normalization on the coordinate point set, and obtain target coordinate data; Using the target coordinate data, an overdetermined method is used to solve multiple sets of plane fitting results for the vehicle surface, and based on the distance from the vehicle surface, an iterative process is performed to determine the target coordinate data with the largest number of retained inliers, and the target coordinate data with the largest number of retained inliers is used to refit the equation for obtaining the final surface fitting result; Based on the equation of the final surface fitting result, the attitude information of the normal vector of the actual vehicle surface and the normal vector of the ideal plane is calculated, and the roll angle around the scanning axis and the pitch angle around the stepping axis are obtained based on the attitude information; Determining a screw adjustment amount and a rotation angle of the ultrasonic scanning fixture based on the roll angle and the pitch angle; The surface of the carrier is divided according to the double-Gaussian bell-shaped division in the two directions of the scanning axis and the stepping axis; During the iterative analysis and processing of the 3D point cloud coordinates, the interior points in the 3D point cloud coordinates are obtained. Based on the number of interior points, the 3D point cloud coordinates of the coefficients of the equation used to fit the final surface fitting result are determined, specifically including: Taking the three-dimensional point cloud coordinates of the vehicle surface with the largest number of inliers as the target point cloud coordinates, determining the deviation of the number of inliers between other point cloud coordinates and the target point cloud coordinates, determining the reference point cloud coordinates among the other point cloud coordinates based on the deviation, obtaining the number of the reference point cloud coordinates, and determining the coordinate credible value of the target point cloud coordinates based on the deviation between the reference point cloud coordinates and the target point cloud coordinates; Taking the reference point cloud coordinates as the target point cloud coordinates, and determining the coordinate credible value of the reference point cloud coordinates based on the number of reference point cloud coordinates when the reference point cloud coordinates are taken as the target point cloud coordinates and the deviation between the reference point cloud coordinates and the target point cloud coordinates; When the sum of the coordinate credibility values ​​of the reference point cloud coordinates is less than a preset threshold, multiple division methods are used, including double Gaussian bell-shaped division and uniform division, and the acquisition of point cloud coordinates continues. The point cloud coordinates with the largest coordinate credibility value among the different division results are used as the three-dimensional point cloud coordinates for fitting the coefficients of the equation of the final surface fitting result. Based on the coordinate credible value of the target point cloud coordinates and the coordinate credible value of the reference point cloud coordinates, the three-dimensional point cloud coordinates of the equation coefficients for fitting the final surface fitting result are determined.

2. The ultrasonic scanning fixture adjustment method based on point cloud data according to claim 1, characterized in that: The three-dimensional point cloud coordinates are is the position of the scanning axis, is the position of the stepper axis, It is the altitude value calculated based on the flight time.

3. The ultrasonic scanning fixture adjustment method based on point cloud data according to claim 1, characterized in that: The value range of N is more than 3.

4. The ultrasonic scanning fixture adjustment method based on point cloud data according to claim 1, characterized in that: Refit the equation to obtain the final plane fitting result, including: Three groups of 3D point cloud coordinates are randomly selected and solved to uniquely determine the fitting result of a vehicle surface; all 3D point cloud coordinates are traversed and the 3D point cloud coordinates whose distance to the vehicle surface is less than a threshold are marked as inliers; multiple iterations are performed to determine the 3D point cloud coordinates of the vehicle surface that retains the largest number of inliers; the fitting result of the vehicle surface that retains the largest number of inliers is used to solve in an overdetermined manner and refit the equation coefficients of the final surface fitting result.

5. The ultrasonic scanning fixture adjustment method based on point cloud data according to claim 1, characterized in that: The normal vector of the actual vehicle surface is obtained by solving the equation of the final surface fitting result.

6. The ultrasonic scanning fixture adjustment method based on point cloud data according to claim 1, characterized in that: The roll angle around the scan axis and the pitch angle around the step axis are determined in the form of Euler angles.

7. The ultrasonic scanning fixture adjustment method based on point cloud data according to claim 1, characterized in that: The calculation formula of the roll angle around the scanning axis is: in is the roll angle around the scan axis, The actual carrier plane Axis normal vector, The actual carrier plane Axis normal vector.

8. The ultrasonic scanning fixture adjustment method based on point cloud data according to claim 1, characterized in that: The screw adjustment amount of the ultrasonic scanning fixture includes the adjustment amount on the x-axis and the y-axis, wherein the calculation formula of the adjustment amount of the x-axis is: , For the vehicle Length in the axial direction.

Images