Method and system for measuring geometric elements of automobile hub

Through three-dimensional visual measurement method, combined with coded pattern projection and point cloud processing, multiple geometric elements of the automobile wheel hub are extracted, which solves the problems of insufficient measurement geometric elements and large errors in the existing technology, and realizes high-precision measurement of wheel hub geometric parameters.

CN119043197AActive Publication Date: 2024-11-29TSINGHUA SHENZHEN INTERNATIONAL GRADUATE SCHOOL

Patent Information

Application Number
CN202411156477.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-22
Publication Date
2024-11-29
Estimated Expiration
2044-08-22

AI Technical Summary

Technical Problem

The existing three-dimensional visual measurement methods measure fewer geometric elements in the measurement of automobile wheel hub geometric elements and are susceptible to light interference and hub texture, resulting in large errors in the measurement results.

Method used

Using a three-dimensional vision-based measurement method, the encoded pattern is projected onto the hub, and the initial point cloud is generated by collecting images, denoising, clustering, plane fitting and scaling projection segmentation are performed, and multiple cover stops and bolts of the hub are extracted. Hole parameters, including diameter, position degree, etc.

Benefits of technology

It realizes the rapid and accurate measurement of multiple hub geometric elements, avoids the impact of image processing complexity and errors during two-dimensional conversion, and improves measurement accuracy and accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119043197A_ABST
    Figure CN119043197A_ABST
Patent Text Reader

Abstract

The invention discloses an automobile hub geometric element measurement method and system, and the method comprises the steps: projecting a coding pattern to a to-be-measured hub through a projector, collecting a hub image through an industrial camera, and generating a hub initial point cloud; denoising the initial point cloud of the hub; clustering the de-noised point clouds, and extracting hub surface point clouds; plane fitting is carried out on the hub surface point cloud; compressing the hub surface point cloud to a fitting plane to obtain a plane point cloud, performing contour extraction on the plane point cloud, and extracting a surface contour point cloud from the hub surface point cloud according to indexes of contour points; the surface contour point cloud is clustered, the maximum cluster is an outer contour, the second large cluster is a cover seam allowance contour, and the other clusters are bolt hole contours; performing zoom projection segmentation on the cover seam allowance contour and the bolt hole contour to obtain a cover seam allowance point cloud and a bolt hole point cloud; and according to the geometrical characteristics of the cover seam allowance and the bolt hole, calculating a cover seam allowance parameter, a bolt hole parameter and the location degree of the bolt hole and the cover seam allowance based on the cover seam allowance point cloud and the bolt hole point cloud.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a method and system for measuring geometrical elements of an automobile wheel hub, and belongs to the technical field of wheel hub detection. Background Art

[0002] The automobile wheel hub is the connecting part between the automobile wheel and the automobile body. It plays an important role in supporting the overall weight of the vehicle, buffering external impact, bearing lateral load, driving force and braking torque. The center hole and bolt hole of the automobile wheel hub are key geometric elements for positioning and connection. Their dimensional accuracy and position accuracy are of great significance to the safe driving of the vehicle. The deformation of the wheel hub will lead to unstable vehicle driving, increased tire wear, and high driving noise. Therefore, accurately and efficiently measuring the key geometric elements of the automobile wheel hub is an important prerequisite for ensuring the performance of the wheel hub and maintaining the safety of vehicle driving.

[0003] At present, the commonly used wheel hub measurement and detection scheme is to calculate the wheel hub parameters through two-dimensional vision. First, the wheel hub image is obtained using an industrial camera, and then the wheel hub geometric parameters are calculated using traditional image processing algorithms. For example, the "Aluminum Alloy Wheel Hub Size Measurement Based on Machine Vision" proposed by Du Lifeng, Zhou Zheng and others obtains the wheel hub contour through the two-dimensional Canny edge detection algorithm, and then fits the contour based on the Hough algorithm and measures the outer contour diameter. The "Research on the Automatic Measurement System of Automobile Wheel Hub Size Based on Machine Vision" proposed by Liu Tao, Luo Yubin and others obtains the wheel hub size data through image processing, feature extraction and data fitting. The "An Intelligent Wheel Hub Model Size Rapid Measurement Method" proposed by Xie Hui, Geng Jiaqin and others processes the wheel hub top view and cross-sectional view, extracts the wheel hub contour, determines the coordinates, and then calculates the wheel hub diameter, width and other dimensions based on the coordinates. Although two-dimensional vision developed earlier, the technology is mature, and the cost is low, it is easily affected by light interference and the texture of the wheel hub itself, which makes the measurement result error large.

[0004] Two-dimensional vision often has certain limitations when facing complex wheel hub structures and high-precision requirements. With the continuous advancement of science and technology, three-dimensional vision methods have brought revolutionary changes to wheel hub detection. It can break through the constraints of traditional measurement and capture the three-dimensional form of the wheel hub in a non-contact, high-precision and high-speed manner, providing a reliable data basis for the optimization of wheel hub manufacturing processes, quality assessment and new product development, leading the field of wheel hub measurement to a new height. In recent years, with the development and popularization of laser radar and three-dimensional scanning technology, this technology has been widely used in various fields. Benefiting from the development of these technologies, measurement technology based on three-dimensional vision has emerged. This technology mainly uses non-contact measurement, which will not cause deformation and damage to the measured surface, and can obtain complete three-dimensional data of the object to be detected. At present, there have been some studies on the measurement of wheel hub geometric elements through three-dimensional vision, but most of them convert the three-dimensional data collected by the scanner into a two-dimensional image, and the measured geometric elements are relatively few. Summary of the invention

[0005] In order to quickly and accurately measure the geometric parameters of the wheel hub, the present invention proposes a measurement method based on 3D vision, which directly processes the 3D point cloud and can quickly and accurately measure 8 geometric elements including the hub cover stop diameter, bolt hole diameter, and position, so as to solve the problem that the existing 3D vision measurement method can measure fewer hub geometric elements.

[0006] To solve the above problems, one aspect of the present invention proposes the following technical solutions:

[0007] A method for measuring geometric elements of a car wheel hub comprises the following steps: S1, projecting a coding pattern onto a wheel hub to be measured and collecting a wheel hub image to generate an initial point cloud of the wheel hub; S2, denoising the initial point cloud of the wheel hub to obtain a denoised point cloud; S3, clustering the denoised point cloud to extract a wheel hub surface point cloud; S4, plane fitting the wheel hub surface point cloud to obtain a fitting plane; S5, compressing the wheel hub surface point cloud to the fitting plane to obtain a plane point cloud, and then extracting a contour from the plane point cloud, and extracting a surface contour point cloud from the wheel hub surface point cloud according to the index of the contour point; S6, using clustering to divide the surface contour point cloud into clusters, wherein the most The largest cluster is the outer contour, the second largest cluster is the cover stop contour, and the remaining clusters are bolt hole contours; S7, scale and project the cover stop contour and the bolt hole contour respectively to obtain the cover stop point cloud and the bolt hole point cloud; S8, according to the geometric features of the cover stop, calculate the cover stop parameters based on the cover stop point cloud, and the cover stop parameters include the cover stop diameter, center hole diameter, cap groove depth and wheel core thickness; according to the geometric features of the bolt hole, calculate the bolt hole parameters based on the bolt hole point cloud, and the bolt hole parameters include the bolt hole countersink diameter, bolt hole diameter and bolt hole depth; calculate the distance between the bolt hole contour and the cover stop contour to obtain the position of the bolt hole and the cover stop.

[0008] On the other hand, the present invention also proposes a vehicle wheel hub geometric element measurement system, comprising: a projector, for projecting a coded pattern onto a wheel hub to be measured; an industrial camera, for acquiring an image of the wheel hub to be measured on which the coded pattern is projected; a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the following steps can be implemented: point cloud reconstruction based on the image of the wheel hub to be measured acquired by the industrial camera to obtain an initial point cloud of the wheel hub; denoising the initial point cloud of the wheel hub to obtain a denoised point cloud; clustering the denoised point cloud to extract a wheel hub surface point cloud; plane fitting the wheel hub surface point cloud to obtain a fitting plane; compressing the wheel hub surface point cloud onto the fitting plane to obtain a plane point cloud, and then extracting contours from the plane point cloud, and extracting the contour points according to the index of the contour points. Extracting a surface contour point cloud from the hub surface point cloud; using clustering processing to divide the surface contour point cloud into clusters, wherein the largest cluster is the outer contour, the second largest cluster is the cover stop contour, and the remaining clusters are bolt hole contours; performing scaling projection segmentation on the cover stop contour and the bolt hole contour, respectively, to obtain a cover stop point cloud and a bolt hole point cloud; according to the geometric features of the cover stop, calculating the cover stop parameters based on the cover stop point cloud, the cover stop parameters include the cover stop diameter, the center hole diameter, the cap groove depth and the wheel core thickness; according to the geometric features of the bolt hole, calculating the bolt hole parameters based on the bolt hole point cloud, the bolt hole parameters include the bolt hole countersink diameter, the bolt hole diameter and the bolt hole depth; calculating the distance between the bolt hole contour and the cover stop contour, and obtaining the positional accuracy of the bolt hole and the cover stop.

[0009] Compared with the prior art, the beneficial effects of the technical solution of the present invention are mainly reflected in: the present invention obtains the three-dimensional point cloud data of the wheel hub by means of three-dimensional vision, and processes the three-dimensional point cloud data through steps S2 to S8 to solve multiple cover stop parameters, multiple bolt hole parameters and the position of the bolt hole and the cover stop from the three-dimensional point cloud data. Compared with the solution of converting the three-dimensional point cloud into a two-dimensional image and then measuring the geometric elements through image processing in the prior art, the present invention avoids the complicated image processing process on the one hand, and avoids the influence of the error in converting to a two-dimensional image on the accuracy of the subsequent measurement of the geometric elements on the other hand, and can measure more wheel hub geometric elements. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is a schematic diagram of a vehicle wheel hub geometric element measurement system according to an embodiment of the present invention;

[0011] Figure 2 is a flow chart of a method for measuring geometric elements of an automobile wheel hub according to an embodiment of the present invention;

[0012] Figure 3 is a schematic diagram of a calculation process of geometric elements of an automobile wheel hub according to an embodiment of the present invention;

[0013] Figure 4 It is a cross-sectional view of a car wheel hub;

[0014] Figure 5 It is a schematic diagram of the depth of the automobile wheel hub cap groove. DETAILED DESCRIPTION

[0015] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods and examples. It should be understood that the purpose of providing the examples is only to illustrate, and not to limit the scope of protection of the present invention.

[0016] The specific embodiment of the present invention provides a method for measuring the geometric elements of an automobile wheel hub. The method first builds a Figure 1 The measurement system shown includes a DLP projector 10, an industrial camera 20 and a computer 30. Based on this system, a preset coding pattern is projected onto the wheel hub 40 to be measured by the DLP projector 10, and then the wheel hub image is collected by the industrial camera 20. Then, the initial point cloud data of the wheel hub is generated by the computer 30 and multiple geometric elements of the wheel hub are calculated after a series of processing. Based on this understanding, the computer 30 should include a computer-readable storage medium, and the storage medium should store a predetermined computer program. When the predetermined computer program is executed by the processor, the following steps should be implemented:

[0017] Perform phase calculation on the wheel image captured by the industrial camera, load the calibration parameters obtained after calibration to complete point cloud reconstruction, and generate the initial point cloud of the wheel;

[0018] De-noising the initial point cloud of the wheel hub to obtain a de-noised point cloud;

[0019] Performing clustering processing on the denoised point cloud to extract the hub surface point cloud;

[0020] Performing plane fitting on the hub surface point cloud to obtain a fitting plane;

[0021] Compressing the hub surface point cloud onto the fitting plane to obtain a plane point cloud, then performing contour extraction on the plane point cloud, and extracting a surface contour point cloud from the hub surface point cloud according to an index of contour points;

[0022] The surface contour point cloud is divided into clusters by clustering processing, wherein the largest cluster is the outer contour, the second largest cluster is the cover stop contour, and the remaining clusters are bolt hole contours;

[0023] Performing scaling projection segmentation on the cover stop contour and the bolt hole contour respectively to obtain a cover stop point cloud and a bolt hole point cloud;

[0024] According to the geometric features of the cover stop, the cover stop parameters are calculated based on the cover stop point cloud, and the cover stop parameters include the cover stop diameter, the center hole diameter, the cap groove depth and the wheel core thickness; according to the geometric features of the bolt hole, the bolt hole parameters are calculated based on the bolt hole point cloud, and the bolt hole parameters include the bolt hole countersunk diameter, the bolt hole diameter and the bolt hole depth; the distance between the bolt hole contour and the cover stop contour is calculated to obtain the positional accuracy of the bolt hole and the cover stop.

[0025] Based on the above system, the specific embodiment of the present invention also proposes a method for measuring the geometric elements of a vehicle wheel hub, referring to Figures 1 to 3 The measuring method comprises the following steps S1 to S8:

[0026] S1. Obtain the initial point cloud: first, install a white non-reflective back plate 50 on the inner side of the wheel core of the wheel hub 40 to be tested, so that the center hole of the wheel hub and the five bolt holes are all in the back plate area, and the back plate plane is tightly fitted with the inner side of the wheel core; then, a predetermined coding pattern (such as stripes) is projected onto the wheel hub 40 to be tested by the DLP projector 10; then, the wheel hub image is collected by the calibrated industrial camera 20; finally, the wheel hub image collected by the industrial camera is phase resolved, and the calibration parameters obtained after calibration are loaded to complete the point cloud reconstruction and generate the initial point cloud of the wheel hub. Among them, the existing technology can be used to reconstruct the point cloud based on the image collected by the camera, which will not be repeated here. In addition, the plane where the lens of the projector is located is parallel to the back plate.

[0027] In the preferred embodiment of the present invention, for deep bolt holes, a back plate and multiple exposure methods are adopted, and a DLP projector and an industrial camera are used to perform three-dimensional measurement of the bolt holes. The principle is to project the bolt holes and collect multiple groups of stripe image sequences under different exposure times, and fuse them into a group of images for solving the three-dimensional point cloud. In addition, for the problem that the height information of the inner wall of the measured object cannot be well obtained due to the deep bolt holes and severe reflection of the inner wall, we indirectly obtain the height information of the bolt holes by reconstructing the white back plate plane, which is convenient for the subsequent calculation of various geometric elements.

[0028] S2, point cloud preprocessing: denoising the initial point cloud of the wheel hub obtained in step S1 to obtain a denoised point cloud.

[0029] Noise, also known as isolated points / outliers / abnormal points, refers to irrelevant or unwanted interference signals or errors in point cloud data. It is usually caused by factors such as the brightness of ambient light, measurement equipment accuracy and system errors, object material and surface texture, and human jitter. Preprocessing is required to remove noise and improve calculation accuracy.

[0030] Statistical filtering can effectively remove those isolated noise points that do not conform to the overall point cloud distribution law, thereby making the point cloud data smoother and more accurate. Therefore, in some embodiments of the present invention, statistical filtering is used to perform denoising on the initial point cloud of the wheel hub to obtain a denoised point cloud sor_filtered_cloud. It should be understood that the present invention may also use other point cloud denoising processing methods, and the present invention is not limited thereto.

[0031] Statistical filtering principle:

[0032] The distance between each point and its n neighboring points forms a Gaussian distribution, whose shape is determined by the mean u and standard deviation σ. Let the coordinates of the i-th point in the point cloud be p i (x i ,y i , z i ), from this point to any neighboring point p n (x n ,y n , z n ) is:

[0033]

[0034] The formula for calculating the average distance between each point and any neighboring point is:

[0035]

[0036] The standard deviation is:

[0037]

[0038] Assume that the standard deviation multiple is std. According to the set k and std thresholds, when the average distance of a point to k points is within the standard range (μ-σ·std, μ+σ·std), the point is retained. If it is not within the range, it is defined as an outlier and deleted.

[0039] S3. Extract wheel hub surface point cloud: Use Euclidean clustering to cluster the denoised point cloud sor_filtered_cloud. This process groups similar contours into clusters according to their spatial proximity, where the largest cluster is the wheel hub surface point cloud surface_cloud.

[0040] S4. Adjust the point cloud pose: The wheel hub surface point cloud surface_cloud is plane-fitted by using the random sampling consensus (RANSAC) algorithm to obtain the fitting plane P_surface. Specifically, the angle between the normal vector of the wheel hub surface point cloud surface_cloud and the Z axis of the world coordinate system is first calculated to obtain the rotation matrix; then the rotation matrix is ​​used to rotate the wheel hub surface point cloud surface_cloud and the denoised point cloud sor_filtered_cloud to adjust the point cloud pose of surface_cloud and sor_filtered_cloud to a state parallel to the XOY plane of the world coordinate system, aiming to reduce the complexity of subsequent processing.

[0041] S5. Extract contour: compress the hub surface point cloud surface_cloud to the fitting plane P_surface to obtain the plane point cloud plane_surface_cloud, then use the alpha-shape algorithm to extract the contour of the plane point cloud plane_surface_cloud, and extract the surface contour point cloud surface_cloud_hull from the hub surface point cloud surface_cloud according to the index of the contour point.

[0042] S6. Automatically identify and annotate the cover stop and bolt hole point clouds: Use Euclidean clustering to divide the surface contour point cloud surface_cloud_hull into clusters, where the largest cluster is the outer contour, the second largest cluster is the cover stop contour center_circle, and the remaining clusters are the bolt hole contours bolt_circle. Since there are multiple bolt holes, the bolt hole contours bolt_circle can be sorted clockwise according to the centroid position in this step.

[0043] S7. Accurately segment the cover stop and bolt hole point clouds: In order to avoid the interference of the inner wall point cloud on the calculation of the hub geometric elements, the present invention embodiment designs a scaling projection segmentation algorithm to accurately segment the cover stop and bolt hole point clouds. The details are as follows:

[0044] S71, reducing and adjusting the lid stop contour center_circle to obtain a new lid stop contour new_center_circle, and then extracting the corresponding point cloud of the projection point of the world coordinate system XOY plane in the denoised point cloud sor_filtered_cloud in the new_center_circle to obtain the lid stop point cloud center_cloud;

[0045] S72. The bolt hole outline bolt_circle is reduced and adjusted to obtain a new bolt hole outline new_bolt_circle, and then the corresponding point cloud of the projection point of the world coordinate system XOY plane in the denoised point cloud sor_filtered_cloud in the new_bolt_circle is extracted to obtain the bolt hole point cloud bolt_cloud.

[0046] The principle of the “reduction adjustment” described in S71 and S72 above is the same, as follows:

[0047] 1) Calculate the center of mass (center of the circle): The center of mass is the average position of all points on the circle, which is obtained by calculating the average value of all points on each coordinate axis;

[0048] 2) Determine the scaling factor: Select a scaling factor based on the radius of the target circle to reduce the size of the circle;

[0049] 3) Create a new point cloud set: prepare a new set to store the points of the reduced circle;

[0050] 4) Reduce the point cloud of the circle: traverse all the points on the circle and reduce them one by one:

[0051] 4.1) For each point, first calculate the position offset of the point relative to the center of mass, that is, the distance from the point to the center of mass;

[0052] 4.2) Multiply this distance by the scaling factor to reduce the distance;

[0053] 4.3) Based on the reduced distance, recalculate the new position of the point. The new position is the center of mass plus the reduced distance vector.

[0054] 4.4) Add the newly calculated position points to the new point set to obtain the point cloud of the reduced circle.

[0055] The “circle” mentioned above refers to the cover stop contour center_circle and several bolt hole contours bolt_circle. The center_circle and several bolt_circles are respectively reduced through the above steps.

[0056] Finally, extract the points of the XOY projection points in the target point cloud (i.e., the denoised point cloud sor_filtered_cloud) in the reduced circle.

[0057] S8. Calculate various parameters of the hub according to the geometric features: Figure 4 and Figure 5In the embodiment of the present invention, the wheel hub geometric parameters to be measured include the cover stop diameter 1, the center hole diameter 2, the cap groove depth 3, the wheel core thickness 4, the bolt hole counterbore diameter 5, the bolt hole diameter 6, the bolt hole depth 7 and the bolt hole and cover stop position.

[0058] S81, calculation of cover stop parameters, including calculation of cover stop diameter 1, center hole diameter 2, cap groove depth 3 and wheel core thickness 4, as follows:

[0059] S811, using Euclidean clustering to cluster the cover stop point cloud center_cloud into two categories: the largest cluster and the remaining clusters, where the largest cluster is the cover stop back plate point cloud center_2, and the remaining clusters are the cover stop countersunk hole surface point cloud center_1;

[0060] S812, using the least square method to perform a circle fitting operation on the lid stop contour center_circle, so as to obtain the lid stop diameter;

[0061] S813, using the alpha-shape algorithm to extract the center hole contour from the cover stop back plate point cloud center_2 and perform a circle fitting operation to obtain the center hole diameter;

[0062] S814, using the RANSAC algorithm, respectively perform plane fitting on the lid stop contour center_circle and the lid stop countersunk hole surface point cloud center_1 and calculate the distance between the two planes to obtain the cap groove depth;

[0063] S815, performing plane fitting on the cover stop contour center_circle and the cover stop back plate point cloud center_2 respectively and calculating the distance between the two planes to obtain the wheel core thickness.

[0064] S82, bolt hole parameter calculation, including calculation of bolt hole countersink diameter, bolt hole diameter and bolt hole depth; as follows:

[0065] S821. Use Euclidean clustering to cluster the bolt hole point cloud bolt_cloud into two categories: the largest cluster and the remaining clusters. The largest cluster is the bolt hole back plate point cloud bolt_2, and the remaining clusters are the bolt hole countersunk surface point cloud bolt_1.

[0066] S822, performing a circle fitting operation on the bolt hole profile bolt_circle based on the least square method to obtain the bolt hole countersink diameter;

[0067] S823, using the alpha-shape algorithm to extract the contour from the bolt hole back plate point cloud bolt_2 and perform a circle fitting operation to obtain the bolt hole diameter;

[0068] S824. Use the RANSAC algorithm to perform plane fitting on the bolt hole back plate point bolt_2 and the bolt hole countersunk surface point cloud bolt_1 respectively, and calculate the distance between the two planes to obtain the bolt hole depth.

[0069] S83. Calculate the position of bolt hole and cover stop:

[0070] Calculate the distance from the center of mass of each bolt_circle to the center of mass of center_circle to obtain the position of the bolt hole and the cover stop. Figure 3 As shown, there are five bolt_circles corresponding to the bolt holes, namely bolt_circle 1, bolt_circle 2, bolt_circle 3, bolt_circle4, and bolt_circle 5.

[0071] In an embodiment of the present invention, the projector and the industrial camera are calibrated before use, and the polynomial model method is used to obtain the calibration parameters of the phase-three-dimensional coordinates for subsequent reconstruction; after the calibration is completed, the encoded fringe image is projected onto the wheel to be measured by the projector; then, the industrial camera performs image acquisition, and uses the phase shift method and the multi-frequency heterodyne method to obtain the absolute phase; finally, the absolute phase is substituted into the calibration parameters obtained before, and a high-quality wheel initial point cloud input_cloud is calculated from the phase information.

[0072] The step of obtaining the calibration parameters of the phase-three-dimensional coordinates using the polynomial model method includes:

[0073] Step 1: Place the calibration plate at any position within the measured range, use an industrial camera to shoot the calibration plate, detect the corner points of the calibration plate, and record the coordinates of each corner point in the imaging plane (u ci , v ci )(i=1,2,3…,n);

[0074] Step 2: Keep the calibration plate still and use a projector to project the sinusoidal fringe pattern onto the calibration plate to obtain the phase diagram at this time and record the phase Φ of each corner point. i (i=1, 2, 3, ..., n);

[0075] Step 3: Adjust the placement of the calibration plate several times, repeat steps 1 and 2 15 times, and obtain multiple sets of corner point coordinates (u ci , v ci ) and the corresponding phase Φ i ;

[0076] Step 4: Calibrate the industrial camera using Zhang Zhengyou's camera calibration method to obtain the internal parameters of the industrial camera and the external parameters corresponding to each calibration plate image. According to the external parameters, the coordinates of the corner points in each calibration plate image in the industrial camera coordinate system can be calculated;

[0077] Step 5: Use the phase of the calibration plate corner point position obtained in step 3 and the calibration plate corner point (u ci , v ci ) in the industrial camera coordinate system (X ci , Y ci , Z ci ) Construct a polynomial

[0078] a 1 X c +a 2 Y c +a 3 Z c +a 4 -a 5 ΦX c -a 6 ΦY c -a 7 ΦZ c -a 8 Φ=0

[0079] By using the least squares method, we can find a 1 -a 8 Eight parameters.

[0080] Phase shift method principle:

[0081] In the standard N-step phase shift method, the grayscale value of the i-th sinusoidal grating image captured by the camera can be expressed as

[0082] I i (x,y)=A(x,y)+B(x,y)cos[φ(x,y)+α i ],i=1,2,3,...,N (3-1)

[0083] Among them, I i (x, y) represents the grayscale value of the point (x, y) in the image captured by the camera at the i-th phase shift; A(x, y) is the background light intensity value at the point; B(x, y) represents its modulation intensity value; φ(x, y) is the main phase value at the point, which is related to the spatial frequency f of the sinusoidal fringes. 0 Related to the current coordinates (x, y); α i is the phase shift increment of the i-th phase shift diagram, which can be expressed by α i =2π(i-1) / N. For a pixel, I iis known, while A, B and φ are unknown. Therefore, when N ≥ 3, the phase principal value φ can be calculated using the least squares algorithm:

[0084]

[0085] Substituting formula (3-1) into formula (3-2) yields:

[0086]

[0087] because and Therefore, the representation of the phase main value can be simplified as:

[0088]

[0089] Principle of multi-frequency heterodyne method:

[0090] In the multi-frequency heterodyne method, it is necessary to use the phase principal values ​​of multiple sets of fringe patterns with different spatial frequencies to obtain the phase principal value of the lower spatial frequency distribution, so that the phase difference can cover the entire measurement range within one cycle. Finally, the absolute phase information is obtained based on the phase difference and its spatial frequency multiple relationship with the original phase principal value. For example, in the 3-frequency heterodyne method, it is necessary to project three sets of fringes with different spatial frequencies, and their corresponding spatial periods are T 1 , T 2 , T 3 , their respective phase principal values ​​are φ 1 ,φ 2 ,φ 3 , the absolute phases after expansion are In practical applications, we only need to obtain the absolute phase corresponding to one frequency, which is taken as the final goal here. In getting Before that, we first need to set the space period to T 1 , T 2 The main phase value and period of the fringe are T 2 , T 3 The principal value of the fringe phase is heterodyned to obtain φ 12 ,φ 23 , then take the same operation to convert φ 12 ,φ 23 Heterodyne to obtain φ 123 According to the principle of heterodyne calculation, φ 12 ,φ 23 ,φ 123 The respective expressions are

[0091]

[0092]

[0093]

[0094] The corresponding spatial periods are

[0095]

[0096] Absolute Phase It is necessary to obtain the corresponding phase main value φ 1 and stripe level k 1 It is deduced that:

[0097]

[0098]

[0099] The Round function is used to round to a nearest integer. It is necessary to obtain the corresponding phase main value φ 12 and stripe level k 12 It is deduced that:

[0100]

[0101]

[0102] Finally, substitute the calibration parameters obtained previously and calculate high-quality 3D point cloud data from the phase information:

[0103] Knowing any point (u c , v c ) and the phase Φ corresponding to that point, as well as the camera’s intrinsic parameters (f cx , f cy , c cx , c cy ), the final three-dimensional coordinates can be obtained by the following formula:

[0104]

[0105] Euclidean clustering principle:

[0106] 1. Select any point p from the input point cloud i (x i ,y i , z i ), calculate the neighborhood points within a preset radius r;

[0107] 2. Calculate each neighborhood point p n (x n ,y n , z n) to this point:

[0108]

[0109] 3. The neighborhood points whose distance is less than the set threshold are clustered into the set Q;

[0110] 4. If the number of elements in Q does not increase, the entire clustering process ends; otherwise, p must be selected from the set Q. i Repeat the above process for points other than point A until the number of elements in Q no longer increases.

[0111] 5. Repeat steps 1-4 until all points are clustered.

[0112] RANSAC (Random Sampling Consensus Algorithm) principle:

[0113] 1. According to mathematical knowledge, at least three points are needed to fit a plane. Therefore, three points are randomly selected from the input point cloud (the point cloud that needs to be fitted by the RANSAC algorithm), and then the plane model parameters A, B, C, and D are calculated according to the following plane equation Ax+By+Cz+D=0;

[0114] 2. Use the remaining data points to test the estimated plane model in the plane equation, calculate the result error, and compare the error with the set error threshold. If it is less than the set threshold, the point is determined as an interior point, and the number of interior points under the parameter model is counted and recorded;

[0115] 3. Continue with steps 1 and 2. If the number of inliers in the current model is greater than the maximum number of inliers that have been saved, update the model parameters. The model parameters that are retained are always the model parameters with the largest number of inliers.

[0116] 4. Repeat steps 1 to 3 and iterate continuously until the iteration threshold is reached, and the model parameters with the largest number of inliers are found. Finally, the model parameters are estimated again using the inliers to obtain the final model parameters.

[0117] Principle of alpha-shape algorithm:

[0118] (1) From the input point set P (i.e., the point cloud to be processed by the alpha-shape algorithm), any point p i (x i ,y i ), in the neighborhood point set R with a preset radius of 2a, take any point p n (x n ,y n ), calculate the point p i (x i ,y i ) and point pn (x n ,y n ) Determine the center p o (x o ,y o );

[0119]

[0120]

[0121] (2) Calculate the distance from other points in the neighborhood point set R to the center of the circle p o (x o ,y o ) distance; if all distances are greater than a, it means there are no other points in the circle, then p i p n is the boundary segment, p i , p n It is the boundary contour point; otherwise it is a non-boundary line segment;

[0122] (3) Repeat the first two steps for the remaining points in the neighborhood point set R until all points in R are judged;

[0123] (4) Repeat the above steps for the remaining points in the point set P until all points in P are judged.

[0124] The method for measuring geometric elements of automobile wheel hubs in the embodiment of the present invention removes noise for bolt holes with high reflectivity and deep depth through statistical filtering. The Euclidean clustering algorithm is used to accurately extract and segment the point cloud of the wheel hub surface. The application of the RANSAC algorithm adjusts the point cloud to a state perpendicular to the Z axis of the world coordinate system, thereby improving the accuracy of the analysis. The alpha-shape and scaled projection segmentation algorithms are used to accurately segment the point clouds of the wheel hub cover stop and each bolt hole, thereby enhancing the calculation accuracy of key parameters.

[0125] The detection accuracy of key geometric elements of the wheel hub reaches 100 microns. The combination of high-precision point cloud data, precise segmentation of cover stoppers and bolt holes, and RANSAC algorithm fitting of planes and circles ensures that the detection accuracy of key geometric elements can reach 100 microns.

[0126] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art of the present invention, several equivalent substitutions or obvious variations can be made without departing from the concept of the present invention, and the performance or use is the same, which should be regarded as belonging to the protection scope of the present invention.

Claims

1. A method for measuring geometric elements of a vehicle wheel hub, characterized in that: The following steps are involved: S1, projecting the coding pattern onto the wheel hub to be tested and collecting the wheel hub image to generate the initial point cloud of the wheel hub; S2, denoising the initial point cloud of the wheel hub to obtain a denoised point cloud; S3, performing clustering processing on the denoised point cloud to extract the hub surface point cloud; S4, performing plane fitting on the hub surface point cloud to obtain a fitting plane; S5, compressing the hub surface point cloud onto the fitting plane to obtain a plane point cloud, then performing contour extraction on the plane point cloud, and extracting a surface contour point cloud from the hub surface point cloud according to an index of contour points; S6, dividing the surface contour point cloud into clusters using clustering processing, wherein the largest cluster is the outer contour, the second largest cluster is the cover stop contour, and the remaining clusters are bolt hole contours; S7, performing scaling projection segmentation on the cover stop contour and the bolt hole contour respectively to obtain a cover stop point cloud and a bolt hole point cloud; S8. According to the geometric features of the cover stop, the cover stop parameters are calculated based on the cover stop point cloud, and the cover stop parameters include the cover stop diameter, the center hole diameter, the cap groove depth and the wheel core thickness; according to the geometric features of the bolt hole, the bolt hole parameters are calculated based on the bolt hole point cloud, and the bolt hole parameters include the bolt hole countersunk diameter, the bolt hole diameter and the bolt hole depth; the distance between the bolt hole contour and the cover stop contour is calculated to obtain the position of the bolt hole and the cover stop.

2. The method for measuring the geometrical elements of a vehicle wheel hub according to claim 1, characterized in that: Before step S1, the method further includes: S0, installing a white non-reflective back plate on the inner side of the wheel core of the wheel hub to be tested, and the plane of the back plate is in contact with the inner side of the wheel core.

3. The method for measuring geometric elements of a vehicle wheel hub according to claim 1, characterized in that: Step S3 includes: clustering the denoised point cloud using Euclidean clustering, and grouping them into clusters according to spatial similarity, wherein the largest cluster is the hub surface point cloud.

4. The method for measuring geometric elements of a vehicle wheel hub according to claim 1, characterized in that: Step S4 performs plane fitting on the hub surface point cloud, specifically including: The angle between the normal vector of the hub surface point cloud and the Z axis of the world coordinate system is calculated using a random sampling consistency algorithm to obtain a rotation matrix; The hub surface point cloud and the denoised point cloud are rotated using the rotation matrix to adjust the postures of the hub surface point cloud and the denoised point cloud to a state parallel to the XOY plane of the world coordinate system.

5. The method for measuring geometric elements of a vehicle hub according to claim 1, characterized in that: Step S7 includes: S71, reducing and adjusting the lid stop contour to obtain a new lid stop contour, and then extracting corresponding point clouds of the projection points of the world coordinate system XOY plane in the denoised point cloud in the new lid stop contour to obtain the lid stop point cloud; S72, shrinking and adjusting the bolt hole contour to obtain a new bolt hole contour, and then extracting the corresponding point cloud of the XOY plane projection point of the world coordinate system in the denoised point cloud in the new bolt hole contour to obtain the bolt hole point cloud.

6. The method for measuring geometric elements of a vehicle wheel hub according to claim 5, characterized in that: In step S71, the cover stop contour is reduced and adjusted, which specifically includes: S711, multiplying the distance between each point on the lid stop contour and the center of mass of the lid stop contour by a preset first scaling factor to obtain a reduced distance vector of each point on the lid stop contour; S712, respectively adding the centroid of the lid stop contour and the reduced distance vectors of each point on the lid stop contour to obtain new positions of each point on the lid stop contour; the set of new positions of each point on the lid stop contour is the new lid stop contour; In step S72, the bolt hole profile is reduced and adjusted, specifically including: S721, multiplying the distance between each point on the bolt hole contour and the center of mass of the bolt hole contour by a preset second scaling factor to obtain a reduced distance vector of each point on the bolt hole contour; S722. Add the centroid of the bolt hole contour and the reduced distance vector of each point on the bolt hole contour respectively to obtain new positions of each point on the bolt hole contour; the set of new positions of each point on the bolt hole contour is the new bolt hole contour.

7. The method for measuring geometric elements of a vehicle wheel hub according to claim 2, characterized in that: In step S8, the lid stopper parameters are calculated based on the lid stopper point cloud according to the lid stopper geometric features, specifically including: Clustering the cover stop point cloud into two categories: the largest cluster and the remaining clusters, wherein the largest cluster is the cover stop back plate point cloud, and the remaining clusters are the cover stop countersunk hole surface point clouds; Performing a circle fitting operation on the lid stop contour to obtain the lid stop diameter; Extracting the center hole contour from the cover stop back plate point cloud and performing a circle fitting operation to obtain the center hole diameter; Performing plane fitting on the cover stop contour and the cover stop countersunk hole surface point cloud respectively and calculating the distance between the two planes to obtain the cap groove depth; Plane fitting is performed on the cover stop contour and the cover stop back plate point cloud respectively, and the distance between the two planes is calculated to obtain the wheel core thickness.

8. The method for measuring geometric elements of a vehicle wheel hub according to claim 2, characterized in that: In step S8, bolt hole parameters are calculated based on the bolt hole point cloud according to the geometric features of the bolt hole, specifically including: Clustering the bolt hole point clouds into two categories: the largest cluster and the remaining clusters, wherein the largest cluster is the bolt hole back plate point cloud, and the remaining clusters are the bolt hole countersunk surface point clouds; Performing a circle fitting operation on the bolt hole profile to obtain the bolt hole counterbore diameter; Extracting the contour of the bolt hole back plate point cloud, and then performing a circle fitting operation on the contour to obtain the bolt hole diameter; Plane fitting is performed on the bolt hole back plate point cloud and the bolt hole countersink surface point cloud respectively, and the distance between the two planes is calculated to obtain the bolt hole depth.

9. The method for measuring geometric elements of a vehicle wheel hub according to claim 2, characterized in that: Step S8 calculates the distance between the bolt hole contour and the cover stop contour to obtain the position of the bolt hole and the cover stop, specifically including: For each bolt hole of the wheel hub to be measured, the distance from the center of mass of the bolt hole contour to the center of mass of the cover stop contour is calculated to obtain the position of each bolt hole and the cover stop.

10. A vehicle wheel hub geometric element measurement system, characterized in that: include: A projector, used for projecting the coding pattern onto the wheel hub to be tested; An industrial camera for acquiring an image of the wheel hub to be tested onto which the coded pattern is projected; A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can implement the following steps: Point cloud reconstruction is performed based on the image of the wheel hub to be tested acquired by the industrial camera to obtain the initial point cloud of the wheel hub; De-noising the initial point cloud of the wheel hub to obtain a de-noised point cloud; Performing clustering processing on the denoised point cloud to extract the hub surface point cloud; Performing plane fitting on the hub surface point cloud to obtain a fitting plane; Compressing the hub surface point cloud onto the fitting plane to obtain a plane point cloud, then performing contour extraction on the plane point cloud, and extracting a surface contour point cloud from the hub surface point cloud according to an index of contour points; The surface contour point cloud is divided into clusters by clustering processing, wherein the largest cluster is the outer contour, the second largest cluster is the cover stop contour, and the remaining clusters are bolt hole contours; Performing scaling projection segmentation on the cover stop contour and the bolt hole contour respectively to obtain a cover stop point cloud and a bolt hole point cloud; According to the geometric features of the cover stop, the cover stop parameters are calculated based on the cover stop point cloud, and the cover stop parameters include the cover stop diameter, the center hole diameter, the cap groove depth and the wheel core thickness; according to the geometric features of the bolt hole, the bolt hole parameters are calculated based on the bolt hole point cloud, and the bolt hole parameters include the bolt hole countersunk diameter, the bolt hole diameter and the bolt hole depth; the distance between the bolt hole contour and the cover stop contour is calculated to obtain the positional accuracy of the bolt hole and the cover stop.

Citation Information

Patent Citations

  • Target detection and classification method and system based on 3D point cloud data

    CN110647835A

  • Magnetic shoe geometric dimension detection method and device

    CN115170640A

  • Surface-mounted inductor geometric parameter measurement method and system

    CN116399241A

  • Material drawing guidance method and system based on three-dimensional back projection, and storage medium

    CN117058233A

  • Indoor structure point cloud segmentation method

    CN117788774A

Cited By

  • Wheel cover detection method and system based on three-dimensional modeling

    CN119533335A

  • A wheel cover detection method and system based on three-dimensional modeling

    CN119533335B

  • Method and system for measuring wheel set size based on wheel three-dimensional point cloud

    CN121521007A