A visual imaging correction method for wheel hub grinding system

Abstract

The present invention belongs to the technical field of robot wheel hub grinding, and specifically refers to a visual imaging correction method for a wheel hub grinding system, the method including fixing equipment, collecting images, identifying wheel hub models, image correction, preliminary grinding and grinding correction. This scheme proposes a visual imaging correction method for a wheel hub grinding system, equipping the wheel hub grinding robot with eyes to help the traditional burr grinding system to achieve automated grinding operations; using a laser line scan camera with stronger resistance to ambient light interference to collect images of the wheel hub surface and rib window edges, correcting the distortion of the wheel hub's three-dimensional point cloud information, setting a semi-mirror between the light source and the laser line scan camera to move the light to the same direction as the laser line scan camera axis; using the method of first remapping the two-dimensional image and then reconstructing the point cloud, the correction image is changed from complex algorithm calculation to pixel position remapping, thereby improving calculation efficiency.

Description

Technical Field

[0001] The invention relates to the technical field of robot wheel hub grinding, and in particular to a visual imaging correction method for a wheel hub grinding system. Background Art

[0002] In the modern industrial manufacturing process, wheels are an important part of tires. On the automated production line of wheels, robot grinding systems have gradually replaced the traditional manual grinding of wheels. Traditional robot grinding requires manual teaching of grinding for new wheels in advance, so that the robot can accurately record the grinding position, which makes the teaching process complicated and time-consuming. In addition, due to the inconsistency of the shapes and sizes of wheels on the production line, the teaching robot cannot adapt to the production of multiple models or mixed models. Computer vision cameras capture the light reflected by objects, so lighting is crucial for correct image processing. The single light source environment of the grinding factory is limited, and the workpiece images obtained by the image acquisition module equipped with the traditional grinding system are usually unusable due to the interference of glare and overexposure on the metal surface of the workpiece. The measurement data points of the wheel are large in volume and high in density, and each pixel position of each image collected by the camera will traverse the correction algorithm at least once, which will result in a particularly large amount of calculation, serious time consumption, and the scanning rhythm cannot keep up with the production rhythm. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a visual imaging correction method for a wheel hub grinding system; the traditional robot grinding requires manual teaching of grinding for the new wheel hub in advance, so that the robot can accurately record the grinding position, which makes the teaching process complicated and time-consuming. In addition, due to the inconsistency of the shapes and sizes of the wheels on the production line, the teaching robot cannot adapt to the production of multiple models or mixed models. This solution proposes a visual imaging correction method for a wheel hub grinding system. With the help of computer vision, the wheel hub grinding robot is equipped with eyes, making the wheel hub grinding process more intelligent and Economical, safe, flexible, and efficient; computer vision cameras capture light reflected by objects, so lighting is crucial for correct image processing. In view of the limited single light source environment in the grinding factory, the workpiece images obtained by the image acquisition module of the traditional grinding system are usually unusable due to the interference of glare and overexposure on the metal surface of the workpiece. This solution uses a laser line scan camera with stronger resistance to ambient light interference to collect images of the hub surface and rib window edge, and correct the distortion of the hub three-dimensional point cloud information. A half-mirror is set between the light source and the laser line scan camera to improve measurement accuracy and stability. The measurement data points of the hub are large in volume and high in density, and each pixel position of each image collected by the camera will traverse the correction algorithm at least once, which will result in a particularly large amount of calculation, serious time consumption, and the scanning rhythm cannot keep up with the production rhythm. This solution adopts the method of remapping the two-dimensional image first and then reconstructing the point cloud, changing the correction image from complex algorithm calculation to pixel position remapping, which greatly saves computing resources.

[0004] The technical solution adopted by the present invention is as follows: The present invention provides a visual imaging correction method for a wheel hub grinding system, the method comprising the following steps:

[0005] Step S1: Fix the device and collect images;

[0006] Step S2: Identify the wheel hub model; collect the CAD model of the wheel hub before casting, record and identify the wheel hub model, if the wheel hub model has not been recorded and it is the first grinding of the wheel hub grinding system, then record the wheel hub model and execute step S3; otherwise jump to step S4;

[0007] Step S3: image correction;

[0008] Step S4: preliminary polishing;

[0009] Step S5: Grinding and correction.

[0010] Furthermore, in step S1, the fixed device collects images, specifically including the following steps:

[0011] Step S11: fix the wheel hub; place the wheel hub on the rotating platform, align the central axis of the wheel hub with the central axis of the rotating platform, and place an encoder on the edge of the rotating platform to measure the rotation state of the wheel hub;

[0012] Step S12: Fix the camera; the hub rib window is the spoke structure of the hub, and the laser line scan camera A is placed on one side of the hub rib window, and the laser line scan camera Z is placed on the other side of the hub rib window;

[0013] Step S13: Optimizing lighting; setting a half mirror between the light source irradiating the hub and the hub, adjusting the positional relationship between the light source and the half mirror, so that the light reflected by the half mirror propagates along the axis direction of the laser line scan camera A and the laser line scan camera Z;

[0014] Step S14: eccentrically install the laser; install the laser at a position offset from the central axis by a preset distance, and record the preset distance as ;

[0015] Step S15: The rotating platform drives the wheel hub to rotate one circle, the laser concentrates energy to emit a laser line to illuminate the surface of the wheel hub, forming reflected light, and the laser line scan camera receives the reflected light. The laser line scan camera A collects the deformation image A of the wheel hub rib window during the rotation of the wheel hub; the laser line scan camera Z collects the deformation image Z of the wheel hub rib window during the rotation of the wheel hub.

[0016] Furthermore, in step S3, the image correction specifically includes the following steps:

[0017] Step S31: Correct the deformed image A in the row direction. The specific steps are as follows:

[0018] Step S311: Calculate the rotation angle, the formula is as follows:

[0019] ;

[0020] ;

[0021] In the formula, It represents the rotation angle between any radius OE of the hub scanned by the laser and the rotation starting radius OC of the hub. It represents the distance between the intersection of any scanned radius OE and the laser line and the intersection of the rotation starting radius OC and the laser line. express The length in the deformed image A, represents the scaling parameter of the laser line along the tangential direction, represents the inverse tangent function, Indicates the preset distance that the laser line is offset from the central axis;

[0022] Step S312: Calculate the line difference, the formula is as follows:

[0023] ;

[0024] In the formula, It represents the difference between the rotation starting radius OC and the laser line on the deformation image A. Indicates the number of pulses measured by the encoder when the wheel hub rotates one circle;

[0025] Step S313: Correct the rows of the deformed image A using the following formula:

[0026] ;

[0027] In the formula, The number of rows where the pixels are located after correction, Indicates the row number where the pixel is located before correction;

[0028] Generate a deformed image B after row correction;

[0029] Step S32: Correct the deformed image B in the column direction. The specific steps are as follows:

[0030] Step S321: Calculate the linkage angle, the formula is as follows:

[0031] ;

[0032] In the formula, represents the linkage angle;

[0033] The intersection point between the straight line of any radius OE and the rotation starting radius OC is defined as point D, the intersection point between the rotation starting radius OC and the laser line is defined as point G, point F on the rotation starting radius OC is located according to the linkage angle, and the distance from point D to point G is defined as , the distance from point G to point F is defined as ;

[0034] Step S322: Calculate , the formula is as follows:

[0035] ;

[0036] Step S323: Correct the columns of the deformed image B using the following formula:

[0037] ;

[0038] In the formula, Indicates the column number of the pixel after calibration. represents the width of the deformed image B, represents the scaling parameter of the laser line along the radial direction;

[0039] Generate a column-rectified wheel image C.

[0040] Further, in step S4, the preliminary grinding specifically includes the following steps:

[0041] Step S41: If the wheel hub model has not been recorded and it is the first time for the wheel hub grinding system to grind, then execute step S42; otherwise, jump to step S43;

[0042] Step S42: creating a correction template; numbering all pixels of the deformed image A in order from left to right and from top to bottom to form an image number matrix, and performing the operation of step S3 on the image number matrix to generate the hub image C following the deformed image A, obtaining a correction template representing the pixel position after correction, and storing the correction template of the hub model;

[0043] Step S43: remapping; applying the correction template of the wheel model to the laser line scan camera A and the laser line scan camera Z and performing a correction operation, performing a polar coordinate transformation on the wheel image C with the central axis as the center point, and obtaining a circular wheel image D;

[0044] Step S44: When the wheel hub reaches the working position of the wheel hub grinding system, the wheel hub grinding system clamps the wheel hub, and according to the two-dimensional circular wheel hub image D, uses an edge detection algorithm to analyze the wheel hub edge contour, generates a grinding trajectory, and performs the wheel hub grinding operation.

[0045] Furthermore, in step S5, the grinding correction specifically includes the following steps:

[0046] Step S51: 3D mapping; transform the circular hub image D after polar coordinate transformation from 2D pixels to 3D point cloud to obtain hub point cloud data. The formula used is as follows:

[0047] ;

[0048] ;

[0049] ;

[0050] In the formula, , , Represent the point cloud coordinates, , , Represent the column, row and pixel value of the pixel respectively. represents the bias parameter of the laser line along the tangential direction, Represents the offset parameter of the laser line along the radial direction;

[0051] Step S52: fusing point clouds; using a transformation matrix to transform the point cloud data of the laser line scan camera A into the point cloud coordinate system of the laser line scan camera Z, and fusing the wheel hub point cloud data of the laser line scan camera A and the laser line scan camera Z and the CAD model of the wheel hub into a complete point cloud image E;

[0052] Step S53: using an edge detection algorithm and a morphological detection algorithm to identify the geometric deformation and remaining burrs of the wheel hub according to the point cloud image E;

[0053] Step S54: Correcting the grinding trajectory according to the geometric deformation of the wheel hub and the volume of the remaining burrs generated by the wheel hub grinding system during the process of clamping the wheel hub.

[0054] The present invention provides a visual imaging correction method for a wheel hub grinding system. The beneficial effects achieved by the present invention using the above scheme are as follows:

[0055] (1) Traditional robot grinding requires manual teaching of grinding for new wheels in advance, so that the robot can accurately record the grinding position, which makes the teaching process complicated and time-consuming. In addition, due to the inconsistency of wheel shapes and sizes on the production line, the teaching robot cannot adapt to the production of multiple models or mixed models. This scheme proposes a visual imaging correction method for the wheel grinding system. With the help of computer vision, the wheel grinding robot is equipped with eyes to provide a low-cost, high-efficiency and high-stability burr grinding solution, which helps the traditional burr grinding system to realize automated grinding operations, making the wheel grinding process more intelligent, economical, safe, flexible and efficient.

[0056] (2) Computer vision cameras capture light reflected from objects, so lighting is crucial for correct image processing. Given the limited environment of a single light source in a grinding factory, the workpiece images obtained by the image acquisition module of a traditional grinding system are usually unusable due to glare and overexposure on the metal surface of the workpiece. This solution uses a laser line scan camera with stronger resistance to ambient light interference to collect images of the hub surface and rib window edges, and correct the distortion of the hub's three-dimensional point cloud information. At the same time, a semi-mirror is set between the light source and the laser line scan camera to move the light to the same direction as the laser line scan camera axis, achieving uniform diffuse light and improving measurement accuracy and stability.

[0057] (3) The measurement data points of the wheel hub are large in volume and high in density. In addition, each pixel position of each image captured by the camera will traverse the correction algorithm at least once, which will result in a particularly large amount of calculation, serious time consumption, and the scanning cycle cannot keep up with the production cycle. This solution adopts the method of first remapping the two-dimensional image and then reconstructing the point cloud, transforming the correction image from complex algorithm calculation to pixel position remapping, which greatly saves computing power resources and improves computing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A schematic flow chart of a visual imaging correction method for a wheel hub grinding system provided by the present invention;

[0059] Figure 2 Schematic diagram of the positional relationship among the wheel hub, laser line scan camera A, laser line scan camera Z and the laser;

[0060] Figure 3 A schematic diagram of the relationship of image correction in step S3;

[0061] Figure 4 is a deformation image A of the wheel hub of Embodiment 9;

[0062] Figure 5 The correction template of embodiment 10;

[0063] Figure 6 is the corrected wheel hub image C of Example 11;

[0064] Figure 7 The circular hub image D of the twelfth embodiment;

[0065] Among them, 1. wheel hub, 2. laser line scan camera A, 3. laser line scan camera Z, 4. laser;

[0066] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0067] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0068] In the description of the present invention, it should be understood that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0069] Example 1, see Figures 1 to 7 The present invention provides a visual imaging correction method for a wheel hub grinding system, the method comprising the following steps:

[0070] Step S1: Fix the device and collect images;

[0071] Step S2: Identify the wheel hub model; collect the CAD model of the wheel hub before casting, record and identify the wheel hub model, if the wheel hub model has not been recorded and it is the first grinding of the wheel hub grinding system, then record the wheel hub model and execute step S3; otherwise jump to step S4;

[0072] Step S3: image correction;

[0073] Step S4: preliminary polishing;

[0074] Step S5: Grinding and correction.

[0075] Example 2, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. In step S1, the fixed device collects images, specifically including the following steps:

[0076] Step S11: fix the wheel hub; place the wheel hub on the rotating platform, align the central axis of the wheel hub with the central axis of the rotating platform, and place an encoder on the edge of the rotating platform to measure the rotation state of the wheel hub;

[0077] Step S12: Fix the camera; the hub rib window is the spoke structure of the hub, and the laser line scan camera A is placed on one side of the hub rib window, and the laser line scan camera Z is placed on the other side of the hub rib window;

[0078] Step S13: Optimizing lighting; setting a half mirror between the light source irradiating the hub and the hub, adjusting the positional relationship between the light source and the half mirror, so that the light reflected by the half mirror propagates along the axis direction of the laser line scan camera A and the laser line scan camera Z;

[0079] Step S14: eccentrically install the laser; install the laser at a position offset from the central axis by a preset distance, and record the preset distance as ;

[0080] Step S15: The rotating platform drives the wheel hub to rotate one circle, the laser concentrates energy to emit a laser line to illuminate the surface of the wheel hub, forming reflected light, and the laser line scan camera receives the reflected light. The laser line scan camera A collects the deformation image A of the wheel hub rib window during the rotation of the wheel hub; the laser line scan camera Z collects the deformation image Z of the wheel hub rib window during the rotation of the wheel hub.

[0081] Example 3, see Figure 1 , Figure 2 and Figure 3 This embodiment is based on the above embodiment. In step S3, the image correction specifically includes the following steps:

[0082] Step S31: Correct the deformed image A in the row direction. The specific steps are as follows:

[0083] Step S311: Calculate the rotation angle, the formula is as follows:

[0084] ;

[0085] ;

[0086] In the formula, It represents the rotation angle between any radius OE of the hub scanned by the laser and the rotation starting radius OC of the hub. It represents the distance between the intersection of any scanned radius OE and the laser line and the intersection of the rotation starting radius OC and the laser line. express The length in the deformed image A, represents the scaling parameter of the laser line along the tangential direction, represents the inverse tangent function, Indicates the preset distance that the laser line is offset from the central axis;

[0087] Step S312: Calculate the line difference, the formula is as follows:

[0088] ;

[0089] In the formula, It represents the difference between the rotation starting radius OC and the laser line on the deformation image A. Indicates the number of pulses measured by the encoder when the wheel hub rotates one circle;

[0090] Step S313: Correct the rows of the deformed image A using the following formula:

[0091] ;

[0092] In the formula, The number of rows where the pixels are located after correction, Indicates the row number where the pixel is located before correction;

[0093] Generate a deformed image B after row correction;

[0094] Step S32: Correct the deformed image B in the column direction. The specific steps are as follows:

[0095] Step S321: Calculate the linkage angle, the formula is as follows:

[0096] ;

[0097] In the formula, represents the linkage angle;

[0098] The intersection point between the straight line of any radius OE and the rotation starting radius OC is defined as point D, the intersection point between the rotation starting radius OC and the laser line is defined as point G, point F on the rotation starting radius OC is located according to the linkage angle, and the distance from point D to point G is defined as , the distance from point G to point F is defined as ;

[0099] Step S322: Calculate , the formula is as follows:

[0100] ;

[0101] Step S323: Correct the columns of the deformed image B using the following formula:

[0102] ;

[0103] In the formula, Indicates the column number of the pixel after calibration. represents the width of the deformed image B, represents the scaling parameter of the laser line along the radial direction;

[0104] Generate a column-rectified wheel image C.

[0105] Example 4, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. In step S4, the preliminary grinding specifically includes the following steps:

[0106] Step S41: If the wheel hub model has not been recorded and it is the first time for the wheel hub grinding system to grind, then execute step S42; otherwise, jump to step S43;

[0107] Step S42: creating a correction template; numbering all pixels of the deformed image A in order from left to right and from top to bottom to form an image number matrix, and performing the operation of step S3 on the image number matrix to generate the hub image C following the deformed image A, obtaining a correction template representing the pixel position after correction, and storing the correction template of the hub model;

[0108] Step S43: remapping; applying the correction template of the wheel model to the laser line scan camera A and the laser line scan camera Z and performing a correction operation, performing a polar coordinate transformation on the wheel image C with the central axis as the center point, and obtaining a circular wheel image D;

[0109] Step S44: When the wheel hub reaches the working position of the wheel hub grinding system, the wheel hub grinding system clamps the wheel hub, and according to the two-dimensional circular wheel hub image D, uses an edge detection algorithm to analyze the wheel hub edge contour, generates a grinding trajectory, and performs the wheel hub grinding operation.

[0110] Example 5, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. In step S5, the grinding correction specifically includes the following steps:

[0111] Step S51: 3D mapping; transform the circular hub image D after polar coordinate transformation from 2D pixels to 3D point cloud to obtain hub point cloud data. The formula used is as follows:

[0112] ;

[0113] ;

[0114] ;

[0115] In the formula, , , Represent the point cloud coordinates, , , Represent the column, row and pixel value of the pixel respectively. represents the bias parameter of the laser line along the tangential direction, Represents the offset parameter of the laser line along the radial direction;

[0116] Step S52: fusing point clouds; using a transformation matrix to transform the point cloud data of the laser line scan camera A into the point cloud coordinate system of the laser line scan camera Z, and fusing the wheel hub point cloud data of the laser line scan camera A and the laser line scan camera Z and the CAD model of the wheel hub into a complete point cloud image E;

[0117] Step S53: using an edge detection algorithm and a morphological detection algorithm to identify the geometric deformation and remaining burrs of the wheel hub according to the point cloud image E;

[0118] Step S54: Correcting the grinding trajectory according to the geometric deformation of the wheel hub and the volume of the remaining burrs generated by the wheel hub grinding system during the process of clamping the wheel hub.

[0119] Example 6, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S311, the = 0.1433085.

[0120] Embodiment 7, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S312, the =20480.

[0121] Embodiment 8, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S14, the =75mm.

[0122] Embodiment 9, see Figure 1 , Figure 2 and Figure 4 This embodiment is based on the above embodiment, and the deformation image A of the wheel hub is as follows: Figure 4 shown.

[0123] Embodiment 10, see Figure 1 , Figure 2 and Figure 5 This embodiment is based on the above embodiment, and the correction template is as follows Figure 5 shown.

[0124] Example 11, see Figure 1 , Figure 2 and Figure 6 This embodiment is based on the above embodiment, and the corrected wheel hub image C is as follows: Figure 6 shown.

[0125] Example 12, see Figure 1 , Figure 2 and Figure 7 This embodiment is based on the above embodiment, and the circular hub image D is as follows: Figure 7 shown.

[0126] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0127] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0128] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A visual imaging correction method for a wheel hub grinding system, characterized in that: The method comprises the following steps: Step S1: Fix the device and collect images; Step S2: Collect the CAD model of the wheel hub before casting, record and identify the wheel hub model, if the wheel hub model has not been recorded and it is the first grinding of the wheel hub grinding system, then record the wheel hub model and execute step S3; otherwise jump to step S4; Step S3: image correction; Step S4: preliminary polishing; Step S5: grinding and correction; In step S1, the fixed device collects images, specifically including the following steps: Step S11: fix the wheel hub; place the wheel hub on the rotating platform, align the central axis of the wheel hub with the central axis of the rotating platform, and place an encoder on the edge of the rotating platform to measure the rotation state of the wheel hub; Step S12: Fix the camera; the hub rib window is the spoke structure of the hub, and the laser line scan camera A is placed on one side of the hub rib window, and the laser line scan camera Z is placed on the other side of the hub rib window; Step S13: Optimizing lighting; setting a half mirror between the light source irradiating the hub and the hub, adjusting the positional relationship between the light source and the half mirror, so that the light reflected by the half mirror propagates along the axis direction of the laser line scan camera A and the laser line scan camera Z; Step S14: eccentrically install the laser; install the laser at a position offset from the central axis by a preset distance, and record the preset distance as ; Step S15: the rotating platform drives the wheel hub to rotate for one circle, the laser concentrates energy to emit a laser line to illuminate the surface of the wheel hub, forming reflected light, and the laser line scan camera receives the reflected light. The laser line scan camera A collects the deformation image A of the wheel hub rib window during the rotation of the wheel hub; the laser line scan camera Z collects the deformation image Z of the wheel hub rib window during the rotation of the wheel hub; In step S3, the image correction specifically includes the following steps: Step S31: Correct the deformed image A in the row direction. The specific steps are as follows: Step S311: Calculate the rotation angle, the formula is as follows: ； ； In the formula, It represents the rotation angle between any radius OE of the hub scanned by the laser and the rotation starting radius OC of the hub. It represents the distance between the intersection of any scanned radius OE and the laser line and the intersection of the rotation starting radius OC and the laser line. express The length in the deformed image A, represents the scaling parameter of the laser line along the tangential direction, represents the inverse tangent function, Indicates the preset distance that the laser line is offset from the central axis; Step S312: Calculate the line difference, the formula is as follows: ； In the formula, It represents the difference between the rotation starting radius OC and the laser line on the deformation image A. Indicates the number of pulses measured by the encoder when the wheel hub rotates one circle; Step S313: Correct the rows of the deformed image A using the following formula: ； In the formula, The number of rows where the pixels are located after correction, Indicates the row number where the pixel is located before correction; Generate a deformed image B after row correction; Step S32: Correct the deformed image B in the column direction. The specific steps are as follows: Step S321: Calculate the linkage angle, the formula is as follows: ； In the formula, represents the linkage angle; The intersection point between the straight line of any radius OE and the rotation starting radius OC is defined as point D, the intersection point between the rotation starting radius OC and the laser line is defined as point G, point F on the rotation starting radius OC is located according to the linkage angle, and the distance from point D to point G is defined as , the distance from point G to point F is defined as ; Step S322: Calculate , the formula is as follows: ； Step S323: Correct the columns of the deformed image B using the following formula: ； In the formula, Indicates the column number of the pixel after calibration. represents the width of the deformed image B, represents the scaling parameter of the laser line along the radial direction; Generate a wheel hub image C after column correction; In step S4, the preliminary grinding specifically includes the following steps: Step S41: If the wheel hub model has not been recorded and it is the first time for the wheel hub grinding system to grind, then execute step S42; otherwise, jump to step S43; Step S42: creating a correction template; numbering all pixels of the deformed image A in order from left to right and from top to bottom to form an image number matrix, and performing the operation of step S3 on the image number matrix to generate the hub image C following the deformed image A, obtaining a correction template representing the pixel position after correction, and storing the correction template of the hub model; Step S43: remapping; applying the correction template of the wheel model to the laser line scan camera A and the laser line scan camera Z and performing a correction operation, performing a polar coordinate transformation on the wheel image C with the central axis as the center point, and obtaining a circular wheel image D; Step S44: When the wheel hub reaches the working position of the wheel hub grinding system, the wheel hub grinding system clamps the wheel hub, and according to the two-dimensional circular wheel hub image D, uses an edge detection algorithm to analyze the wheel hub edge contour, generates a grinding trajectory, and performs the wheel hub grinding operation.

2. The visual imaging correction method for a wheel hub grinding system according to claim 1, characterized in that: In step S5, the grinding correction specifically includes the following steps: Step S51: 3D mapping; transform the circular hub image D after polar coordinate transformation from 2D pixels to 3D point cloud to obtain hub point cloud data. The formula used is as follows: ； ； ； In the formula, , , Respectively represent the point cloud coordinates, , , Represent the column, row and pixel value of the pixel respectively. represents the bias parameter of the laser line along the tangential direction, Represents the offset parameter of the laser line along the radial direction; Step S52: fusing point clouds; using a transformation matrix to transform the point cloud data of the laser line scan camera A into the point cloud coordinate system of the laser line scan camera Z, and fusing the wheel hub point cloud data of the laser line scan camera A and the laser line scan camera Z and the CAD model of the wheel hub into a complete point cloud image E; Step S53: using an edge detection algorithm and a morphological detection algorithm to identify the geometric deformation and remaining burrs of the wheel hub according to the point cloud image E; Step S54: Correcting the grinding trajectory according to the geometric deformation of the wheel hub and the volume of the remaining burrs generated by the wheel hub grinding system during the process of clamping the wheel hub.

Images