Alignment device, calibration and alignment method for display panel detection equipment

Abstract

The present invention discloses an alignment device, calibration and alignment method for display panel detection equipment, comprising S1: calibrating the installation position of a first camera and the origin position of a second camera; S2: calculating the position offset of a moving platform relative to the second camera when the rotating table is in an initial position according to the offset of one of the reference holes on the detection fixture observed by the first camera relative to the center of the field of view of the first camera before and after the rotating table rotates; S3: calculating the angle between the X-axis direction of the rotating table and the X-axis direction of the device coordinate system according to the offset of one of the reference holes on the detection fixture observed by the first camera relative to the center of the field of view of the first camera before and after the rotating table moves along the X-axis; S4: calculating the posture of the detection fixture relative to the rotating table when the rotating table is in the initial position according to the offset of two reference holes observed by the first camera relative to the center of the field of view of the first camera; the calibration method improves the alignment accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of flat panel display, and in particular to an alignment device, a calibration and alignment method for a display panel detection device. Background Art

[0002] During the production process of a display panel, it is necessary to detect the densely distributed circuits on the panel, determine their on-off status at an early stage, and then decide whether to perform repair or discard. This can avoid bringing defects that occur during the process into subsequent production links, resulting in waste of production capacity and increased costs. The accuracy and speed of the detection device affect the efficiency and profitability of the production line.

[0003] In the design of a display panel, multiple detection areas are usually uniformly arranged near the edge of the panel, and each detection area has a large number of densely distributed test points. During detection, whether the detection head of the detection device can accurately align with the detection area of the product to be tested largely determines the accuracy of the detection. In the production of detection devices, manufacturing errors inevitably exist, and there is also a range of movement and adjustment during component assembly. At the current technical level, these errors and variable factors will cause the detection head of the detection device to not accurately align with the detection area. Summary of the Invention

[0004] Based on the technical problems existing in the background art, the present invention proposes an alignment device, a calibration and alignment method for a display panel detection device, which solves the problem of accurate alignment of multiple probe needles and multiple test points in a test area during a single test, has high detection flexibility and high detection efficiency, and improves the credibility of the detection results.

[0005] A calibration device for a display panel detection device proposed by the present invention includes a first camera and a gantry that can move along the X direction. One or more detection unit groups are arranged on the gantry. Each detection unit group includes a rotary table that can move along the Y axis and can move up and down, and a second camera. A detection jig is arranged on the moving platform of the rotary table. The moving platform of the rotary table can move along the X direction and can also rotate around the Z axis. Two reference holes and two or more probe needles are arranged on the detection jig.

[0006] A calibration method for a display panel detection device includes the following steps:

[0007] S1: Calibrate the installation position of the first camera and the origin position of the second camera through a marking pattern C;

[0008] S2: According to the offset of one of the reference holes on the detection jig relative to the center of the field of view of the first camera before and after the rotation of the rotary table observed by the first camera, calculate the position offset of the moving platform of the rotary table relative to the second camera when the rotary table is in the initial position;

[0009] S3: According to the detection by the first camera, detect the offset of one of the reference holes on the detection fixture relative to the center of the field of view of the first camera before and after the rotary table moves along the X-axis, and calculate the angle between the X-axis direction of the rotary table and the X-axis direction of the equipment coordinate system;

[0010] S4: According to the detection by the first camera of the offsets of the two reference holes relative to the center of the field of view of the first camera, calculate the pose of the detection fixture relative to the rotary table when the rotary table is in the initial position;

[0011] S5: Repeat steps S1 to S4 to calculate the poses of other detection fixtures relative to the rotary table and the pose of the rotary table relative to the second camera.

[0012] Further, in step S1: Calibrate the installation position of the first camera and the origin position of the second camera 21 through the marking pattern C. The calibration of the installation position of the first camera is specifically as follows:

[0013] Take the center of the marking pattern C as the origin, the movement direction of the gantry X-axis as the X-axis, and the movement direction of the Y11 axis of the detection unit group as the Y-axis to establish an equipment coordinate system;

[0014] Adjust the installation position of the first camera to make the center of the field of view of the first camera close to the marking pattern C;

[0015] Determine the offset of the center of the field of view of the first camera relative to the marking pattern C, denoted as (Aoffsx, Aoffsy).

[0016] Further, in step S1: Calibrate the installation position of the first camera and the origin position of the second camera 21 through the marking pattern C. The calibration of the origin position of the second camera is specifically as follows:

[0017] The gantry X-axis and the Y11 axis of the detection unit group return to the reference point. Set the offset of the reference point of the gantry X-axis relative to the X-axis of the equipment coordinate system at this time as X RefOffs0 , and the offset of the reference point of the Y11 of the detection unit group relative to the Y-axis of the equipment coordinate system as Y11 RefOffs0 ;

[0018] Move the detection unit group to align the center of the field of view of the second camera with the marking pattern C, calibrate the origin position of the second camera, and record the coordinate of the gantry X-axis at this time as x a , and the coordinate of the Y11 axis of the detection unit group as y b ;

[0019] Modify the offset of the reference point of the gantry X-axis relative to the X-axis of the equipment coordinate system to (X RefOffs0 - x a ), and modify the offset of the reference point of the Y11 axis of the detection unit group relative to the Y-axis of the equipment coordinate system to (Y11 RefOffs0 - yb ), the coordinates (X RefOffs0 -x a , Y11 RefOffs0 -y b ) are used as the reference point positions after the second camera calibration.

[0020] Further, in step S2: According to the offset of one of the reference holes on the detection jig relative to the center of the field of view of the first camera before and after the rotation of the rotary table, calculate the position offset of the moving platform relative to the second camera when the rotary table is in the initial position, which specifically includes:

[0021] The rotary table returns to the reference position, and the X-axis of the gantry and the Y11-axis of the detection unit group return to the reference point again;

[0022] Move the X-axis of the gantry and the Y11-axis of the detection unit group so that the reference hole at the lower end of the detection jig appears in the field of view of the first camera;

[0023] Record the offset of the reference hole relative to the center of the field of view of the first camera at this time, denoted as (h1x, h1y);

[0024] Record the coordinates of the X-axis of the gantry and the Y11-axis of the detection unit group at this time, denoted as b1x, b1y;

[0025] Calculate the offset of the reference hole 231 relative to the second camera 21 as (off1x, off1y) according to (Aoffsx, Aoffsy), (h1x, h1y), b1x, b1y, where off1x = h1x + Aoffsx - b1x, off1y = h1y + Aoffsy - b1y;

[0026] Operate the moving platform of the rotary table to rotate through an angle θ;

[0027] Move the X-axis of the gantry and the Y11-axis of the detection unit group to make the same reference hole at the lower end of the detection jig appear in the field of view of the first camera again;

[0028] Record the offset of the reference hole relative to the center of the field of view of the first camera at this time, denoted as (h2x, h2y);

[0029] Record the coordinates of the X-axis of the gantry and the Y11-axis of the detection unit group at this time, denoted as b2x, b2y;

[0030] Calculate the offset of the reference hole relative to the second camera as (off2x, off2y) according to (Aoffsx, Aoffsy), (h2x, h2y), b2x, b2y, where off2x = h2x + Aoffsx - b2x, off2y = h2y + Aoffsy - b2y;

[0031] Based on (off1x, off1y), (off2x, off2y), and θ, calculate the position offset (X UvwCam , Y UvwCam ) of the moving platform relative to the second camera when the rotary table is in the initial position, where X UvwCam = avgx - Δy / (2 × tan(θ / 2)), Y UvwCam = avgy + Δx / (2 × tan(θ / 2)), Δx = off2x - off1x, Δy = off2y - off1y, avgx = (off1x + off2x) / 2, avgy = (off1y + off2y) / 2.

[0032] Furthermore, in step S3: According to the offset of one of the reference holes on the detection jig relative to the center of the field of view of the first camera before and after the rotary table moves along the X-axis, calculate the angle between the X-axis direction of the rotary table and the X-axis direction of the device coordinate system, specifically including:

[0033] The rotary table returns to the reference position, and the moving platform of the rotary table is operated to move a certain distance in the positive X direction relative to the fixed platform of the rotary table;

[0034] Move the X-axis of the gantry and the Y11-axis of the detection unit group so that the same reference hole at the lower end of the detection jig appears in the field of view of the first camera;

[0035] Record the offset of the reference hole relative to the center of the field of view of the first camera at this time, denoted as (h3x, h3y);

[0036] Record the coordinates of the X-axis of the gantry and the Y11-axis of the detection unit group at this time, denoted as b3x, b3y;

[0037] Calculate the offset of the reference hole 231 relative to the second camera 21 as (off3x, off3y) based on (Aoffsx, Aoffsy), (h3x, h3y), b3x, and b3y, where off3x = h3x + Aoffsx - b3x, off3y = h3y + Aoffsy - b3y;

[0038] Calculate the angle Th between the X-axis direction of the rotary table and the X-axis direction of the device coordinate system based on (off1x, off1y) and (off3x, off3y) UvwCam , where Th UvwCam = atan2((off3y - off1y), (off3x - off1x)).

[0039] Further, in step S4: calculating the pose of the detection jig relative to the rotary table when the rotary table is at the initial position according to the offsets of two reference holes relative to the center of the field of view of the first camera, specifically including:

[0040] The rotary table returns to the reference position, and the gantry X-axis and the detection unit group Y11 are operated so that another reference hole at the lower end of the jig appears in the field of view of the first camera;

[0041] Record the offset of this reference hole relative to the center of the field of view of the first camera at this time, denoted as (h4x, h4y);

[0042] Record the coordinates of the gantry X-axis and the Y11 axis of the detection unit group at this time, denoted as b4x, b4y;

[0043] Calculate the offset of this reference hole relative to the second camera as (off4x, off4y) according to (Aoffsx, Aoffsy), (h4x, h4y), b4x, b4y, where off4x = h4x + Aoffsx - b4x, off4y = h4y + Aoffsy - b4y;

[0044] According to (off1x, off1y), (off4x, off4y), (X UvwCam , Y UvwCam ) calculate the offset (X JigUvw , Y JigUvw ) of the detection jig 23 relative to the rotary table, where, X JigUvw = off1x / 2 + off4x / 2 - X UvwCam , Y JigUvw = off1y / 2 + off4y / 2 - Y UvwCam ;

[0045] According to (off1x, off1y), (off4x, off4y), Th UvwCam calculate the angle TH JigUvw that the detection jig 23 rotates relative to the rotary table, where, TH JigUvw = atan2((off4y - off1y), (off4x - off1x)) - Th UvwCam .

[0046] A method for aligning a display panel detection device, characterized by including the following steps:

[0047] Observing the GMark coordinates of the substrate to be tested after loading through the second camera to determine the coordinates of multiple test points in a test area;

[0048] Determine the test area coordinate system X according to the coordinates of multiple test pointsC O C Y C The pose (X JigDev , Y JigDev , TH JigDev ) of OY is used as the target pose of the detection fixture (23) for the coordinate system pose (X JigDev , Y JigDev , TH JigDev ) of the test area coordinate system;

[0049] Calculate the X-axis coordinate of the gantry, the Y11-axis coordinate of the detection unit group, and the pose of the rotary table according to the calibration method and the target pose of the detection fixture.

[0050] Further, in calculating the X-axis coordinate of the gantry, the Y11-axis coordinate of the detection unit group, and the pose of the rotary table according to the calibration method and the target pose of the detection fixture (23), it specifically includes: Assuming that the X-axis of the gantry is determined, from the offset and rotation angle (X JigUvw , Y JigUvw , TH JigUvw ) of the detection fixture relative to the rotary table, the offset and rotation angle (X UvwCam , Y UvwCam , TH UvwCam ) of the rotary table relative to the second camera, and the target pose (X JigDev , Y JigDev , TH JigDev ) of the detection fixture, inversely solve the pose (X Uvw , θ) of the rotary table and the Y11-axis coordinate of the detection unit group.

[0051] Further, in calculating the X-axis coordinate of the gantry, the Y11-axis coordinate of the detection unit group, and the pose of the rotary table according to the calibration method and the target pose of the detection fixture, it specifically includes:

[0052] Assuming that the rotary table only rotates, from the offset (X JigUvw , Y JigUvw ) of the detection fixture relative to the rotary table, the target pose (X JigDev , Y JigDev , TH JigDev ) of the detection fixture, and the pose (X UvwCam , Y UvwCam , TH UvwCam ) of the moving platform relative to the second camera when the rotary table (22) is in the initial position, inversely solve the X-axis coordinate of the gantry, the Y11-axis coordinate of the detection unit group, and the rotation axis coordinate of the rotary table.

[0053] Further, the specific formulas for inversely solving the pose (X Uvw , θ) of the rotary table and the Y11-axis coordinate of the detection unit group are as follows:

[0054] θ = TH JigDev -TH JigUvw -TH UvwCam ;

[0055] X JigDev = X + X JigCam ;

[0056] X JigCam = X JigUvw cosθ - Y JigUvw sinθ + X Uvw cosTH UvwCam + X UvwCam ;

[0057] Y JigCam = X JigUvw sinθ + Y JigUvw cosθ + X Uvw sinTH UvwCam + Y UvwCam ;

[0058] Y11 = Y JigDev -Y JigCam ;

[0059] Wherein, θ is the rotation angle of the rotary table, X represents the X coordinate of the X-axis of the gantry, X Uvw represents the moving target in the X direction of the rotary table itself, and Y11 represents the Y coordinate of the Y11-axis of the detection unit group.

[0060] Furthermore, the specific formulas for calculating the X-axis coordinate of the gantry, the Y11-axis coordinate of the detection unit group, and the rotation axis coordinate of the rotary table are as follows:

[0061] θ = TH JigDev -TH JigUvw -TH UvwCam

[0062] X JigCam = X JigUvw cosθ - Y JigUvw sinθ + X UvwCam

[0063] Y JigCam = X JigUvw sinθ + Y JigUvw cosθ + Y UvwCam

[0064] Y11 = Y JigDev -Y JigCam

[0065] X = X JigDev -X JigCam

[0066] Among them, θ is the rotation angle of the rotary table, X represents the X coordinate of the X-axis of the gantry, Y11 represents the Y coordinate of the Y11-axis of the detection unit group, (X JigCam , Y JigCam ) represents the offset of the detection jig relative to the second camera in the device coordinate system.

[0067] Further, for multiple test areas on the substrate to be tested with approximate X coordinates, according to the method of inverse-solving the X-axis coordinate of the gantry, the Y11-axis coordinate of the detection unit group, and the rotation axis coordinate of the rotary table, multiple X coordinate values of the gantry are calculated, and the target position of the gantry is calculated by evenly calculating the multiple X coordinate values. Then, the pose (X Uvw , θ) of the rotary table and the Y11-axis coordinate of the detection unit group are inversely solved, and multiple sets of poses (X Uvw , θ) of the rotary table and the Y11-axis coordinate of the detection unit group are calculated.

[0068] Move the gantry to the target position of the gantry, and then move the rotary table and the detection unit group to the poses (X Uvw , θ) of each group of rotary tables and the Y11-axis coordinate of the detection unit group in sequence, so that the detection jig aligns with multiple test areas in sequence.

[0069] The advantages of the alignment device, calibration and alignment method of the display panel detection device provided by the present invention are as follows: In the structure of the present invention, an alignment device, calibration and alignment method of the display panel detection device are provided. By calibrating, the position relationship between the detection jig and the rotary table, and the position relationship between the rotary table and the second camera are obtained. When the device works, combined with the pose of the substrate to be tested observed by the camera, when determining that multiple probes on the detection jig align with multiple test points in the test area, the coordinates of the detection unit group and the coordinates of the rotary table are determined, and the movement of each unit group is controlled with this as the target, so that multiple probes are accurately aligned with the test area; thus, through this detection device, the problem of accurate alignment of multiple probes and multiple test points in multiple test areas during a single test is solved. It is applicable to test equipment with multi-needle detection jigs, has high detection flexibility and high detection efficiency, and is applicable to the scenario of mass production of products on an assembly line. Description of the Drawings

[0070] Figure 1 Flow schematic diagram of the calibration method of the display panel detection device of the present invention;

[0071] Figure 2 Structural schematic diagram of the calibration device of the display panel detection device;

[0072] Figure 3 Parameters and structural schematic diagram of the calibration of the rotary table and the detection jig;

[0073] Figure 4 Schematic diagram of parameters calibrated for the rotary table

[0074] Figure 5 Schematic structural diagram of the fixture coordinate system

[0075] Figure 6 Schematic diagram of parameters and structure of the pose in the test area

[0076] Among them, 1-gantry, 2-detection unit group, 3-substrate to be tested, 4-X11 linear motor, 5-X12 linear motor, 21-second camera, 22-rotary table, 23-detection fixture, 24-Y11 substrate, 25-Z11 substrate, 231-reference hole, 232-probe Specific implementation mode

[0077] Next, the technical solution of the present invention will be described in detail through specific embodiments. Many specific details are set forth in the following description in order to fully understand the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific implementations disclosed below.

[0078] First, as Figures 1 to 6 shown, an alignment device of a display panel detection device includes a frame and a carrier table provided on the frame. A first camera and a transparent medium with a marking pattern C are fixedly installed on one side of the carrier table. The first camera is fixedly installed below the marking pattern C for observing the marking pattern C upward. A gantry 1 is provided on the carrier table. The gantry 1 is driven by an X11 linear motor 4 and an X12 linear motor 5 to move in the X direction. Among them, the gantry 1 is fixed on a slider, and the slider moves along a slide rail to realize the movement of the gantry 1. One or more detection unit groups 2 are provided on the gantry 1. Each detection unit group 2 is connected to the gantry 1 through a Y11 substrate 24 and a Z11 substrate 25. The Z11 substrate 25 is provided on the Y11 substrate 24, and the Y11 substrate 24 is provided on the gantry 1. Among them, the gantry 1 can move horizontally along the carrier table, the Y11 substrate 24 moves longitudinally along the gantry 1, the Z11 substrate 25 can move up and down relative to the Y11 substrate 24, the longitudinal movement direction of the detection unit group 2 corresponds to the movement direction of the Y11 substrate 24, and the detection unit group 2 realizes longitudinal and up-and-down movements through the Y11 substrate 24 and the Z11 substrate 25, and realizes lateral movement through the X-axis movement of the gantry 1. Therefore, the detection unit group 2 can realize three-dimensional movement.

[0079] Each detection unit group 2 includes a second camera 21 and a rotary worktable 22. The second camera 21 is fixed on the Z11 substrate 25, and the fixed platform of the rotary worktable 22 is fixed on the Z11 substrate 25. In this way, on the one hand, the second camera 21 can reach any (X, Y) position in the horizontal direction, and when the thickness of the detection substrate changes, it can also move up and down to achieve clear imaging. The moving platform of the rotary worktable 22 can rotate around the Z axis relative to the fixed platform or translate along its own X axis. Therefore, the moving platform of the rotary platform can rotate around the Z axis of the equipment coordinate system and move along its own X axis relative to the fixed platform. A detection jig 23 is arranged on the moving platform of the rotary worktable 22. Therefore, the detection jig 23 can rotate around the Z axis of the equipment coordinate system and move along the X axis of the rotary worktable through the movement of the moving platform of the rotary platform relative to the fixed platform.

[0080] It should be noted that the translation of the detection jig 23 in the horizontal plane can be achieved by the movement of the X axis and the Y11 axis of the gantry 1, or by the movement of the X axis and the Y11 axis of the rotary worktable 22. However, the latter is limited by the stroke of the X axis of the rotary worktable 22 and has a small range, which can be used as the basis for fine adjustment, and the former can be used as the basis for rough adjustment. The equipment coordinate system is established with the center of the marking pattern C as the origin, the movement direction of the X axis of the gantry 1 as the X axis, and the movement direction of the Y11 axis of the detection unit group 2 as the Y axis.

[0081] Two reference holes 231 and two or more probe needles 232 are arranged on the detection jig 23. There is an accurate positional relationship between the probe needles 232 and the reference holes 231. The relative positional relationship between the multiple probe needles 232 is the same as the relative positional relationship between multiple test points in a test area on the substrate 3 to be tested. According to different products to be detected, the number and distribution of the probe needles 232 on the detection jig 23 are also different.

[0082] In addition, it should be noted that both the second camera 21 and the first camera have a calibration process after installation (before use), which is a common technology of industrial cameras. Its purposes are: ① To determine the physical scale size corresponding to one pixel of the camera. ② To determine the included angle between the camera's field of view direction and the equipment's movement direction.

[0083] When the camera (corresponding to the first camera and the second camera 21) is installed with an inclination, the conversion relationship between the pixel coordinates (a, b) and the device coordinates (x, y) of the image seen by the camera is not only a proportional scaling relationship, but also the angle needs to be considered. Even if the camera is installed with an inclination, after camera calibration, the accurate device coordinates can always be calculated based on the observed pixel coordinates. From the perspective of coordinate calculation effect, it is the same as if the camera's field of view has no inclination. Therefore, in this embodiment, it is considered that the field of view directions of the first camera and the second camera 21 after calibration are consistent with the direction of the device coordinate system, and at the same time, the scaling ratio from pixels to the true scale is also determined. In other words, it can be considered that after the camera is calibrated, any target can be observed through the camera to obtain its offset relative to the center of the camera's field of view in the device coordinate system.

[0084] In this embodiment, there are the following several setting methods: (1) Set a gantry 1, on which a detection unit group 2 is set, and on the detection unit group 2, a detection jig 23 is set. The detection jig 23 can contact a test area for detection; (2) Set a gantry 1, on which at least two detection unit groups 2 are set. Specifically, it is described with two detection unit groups 2 set on the gantry 1: Each of the two detection unit groups 2 is equipped with a detection jig 23. The two detection jigs 23 can be simultaneously aligned with two test areas respectively to improve the detection efficiency. At the same time, the two detection unit groups 2 can cooperate with each other to complete the work of more complex circuits. At this time, the on-off detection is not limited to the on-off between any two test points touched by any two probe needles 232 in the same detection jig 23, but can also detect the on-off between any two test points touched by all the probe needles 232 of the two detection jigs 23. According to the product to be tested, the two detection jigs 23 can be the same or different; (3) Set two or more gantries 1, on each of which at least one detection unit group 2 is set, and each detection unit group 2 is equipped with a detection jig 23, which can improve the detection efficiency. The detection jigs 23 on two or more gantries 1 can also contact two or more test areas simultaneously for cooperative detection to achieve the on-off detection between test points at a long distance.

[0085] This detection device solves the problem of accurately aligning multiple probe needles 232 and multiple test points in the test area during a single test. It is applicable to test devices with multi-needle detection jigs 23, with high detection flexibility and high detection efficiency, and is suitable for the scenario of mass production of products on a production line.

[0086] This detection device needs to be calibrated before use to overcome the influence of manufacturing and assembly errors, and realizes the simultaneous accurate alignment of multiple probe needles 232 to the test area, ensuring the alignment accuracy between the detection unit group 2 and the product to be detected during the detection process. Specifically as follows:

[0087] Such as Figure 1As shown in the figure, a calibration method for a display panel detection device proposed by the present invention includes the following steps S1 to S5:

[0088] S1: Calibrate the installation position of the first camera and the origin position of the second camera 21 through the marking pattern C;

[0089] The calibration of the installation position of the first camera is specifically as follows: Adjust the installation position of the first camera so that the field of view center of the first camera approaches the marking pattern C; Determine the offset of the field of view center of the first camera relative to the marking pattern C, denoted as (Aoffsx, Aoffsy).

[0090] The calibration of the installation position of the second camera 21 is specifically as follows:

[0091] S11: The X-axis of the gantry 1 and the Y11-axis of the detection unit group 2 return to the reference point, and record the offset of the reference point of the X-axis of the gantry 1 relative to the X-axis of the device coordinate system at this time as X RefOffs0 , and record the reference point offset Y11 of the Y11-axis of the detection unit group 2 RefOffs0 ;

[0092] Here, the reference point offset is a device parameter, indicating the offset of the reference point position relative to the origin position of the axis in the device. The "return to the reference point" here represents a working mode of the device. In this mode, a certain axis moves in a preset manner. When encountering a preset signal related to a certain fixed position, the movement stops, and the coordinate of this axis is changed to the reference point offset parameter value of this axis.

[0093] S12: Move the detection unit group 2 so that the field of view center of the second camera 21 aligns with the marking pattern C, calibrate the origin position of the second camera 21, and record the coordinate of the X-axis of the gantry 1 at this time as x a , and the Y11-axis coordinate of the detection unit group 2 as y b ;

[0094] Among them, the X-axis of the gantry 1 and the Y11-axis of the detection unit group 2 are both actual single axes. Therefore, their coordinates are only the coordinates of this axis, and only one component is required, corresponding to x a 、y b .

[0095] S13: Modify the reference point offset of the X-axis of the gantry 1 to (X RefOffs0 -x a ), modify the reference point offset of the Y11-axis of the detection unit group 2 to (Y11 RefOffs0 -y b ), and the X-axis of the gantry 1 and the Y11-axis of the detection unit group 2 perform the operation of returning to the reference point.

[0096] After the reference point offset calibration is performed through steps S11 to S13, the X-axis coordinate of the gantry 1 and the coordinate value of the Y11 axis of the detection unit group 2 can accurately represent the coordinate value of the center of the field of view of the second camera 21 in the device coordinate system.

[0097] S2: Calculate the position offset of the moving platform relative to the second camera 21 when the rotary table 22 is in the initial position according to the offset amount of one of the reference holes 231 on the detection jig 23 relative to the center of the field of view of the first camera before and after the rotary table 22 rotates and the coordinates of the second camera 21.

[0098] S3: Calculate the angle between the X-axis direction of the rotary table 22 and the X-axis direction of the device coordinate system according to the offset amount of one of the reference holes 231 on the detection jig 23 relative to the center of the field of view of the first camera before and after the rotary table 22 moves along the X-axis and the coordinates of the second camera 21.

[0099] S4: Calculate the pose of the detection jig 23 relative to the rotary table 22 when the rotary table 22 is in the initial position according to the offset amounts of the two reference holes 231 relative to the center of the field of view of the first camera and the coordinates of the second camera 21.

[0100] S5: Repeat steps S1 to S4 to calculate the poses of other detection jigs 23 relative to the rotary table 22 and the pose of the rotary table 22 relative to the second camera 21.

[0101] There are multiple test areas on the substrate 3 to be tested, and each test area has multiple test points. There is a definite relative position relationship between the multiple test points. The detection jig 23 has multiple probing needles 232. When the detection jig 23 drops, the probing needles 232 of the same detection jig 23 are in contact with the multiple test points of the test area simultaneously. Through steps S1 to S5, a method for calibrating the pose of the detection device is provided. Based on this calibration result, during production, according to the pose of the substrate after loading, the poses of the rotary table 22 and the detection unit group 2 can be accurately calculated, enabling the multiple probing needles 232 of the jig to be aligned with the multiple test points of the test area simultaneously, improving the alignment accuracy between the detection unit group 2 and the product to be detected during the detection process.

[0102] In addition, there are multiple probing needles 232 on one detection jig 23, which improves the detection efficiency. By adding the rotary table 22, the multiple probing needles 232 of the detection jig 23 can be accurately aligned with the multiple test points of the test area simultaneously. A method for calibrating the installation positions of the rotary table 22 and the detection jig 23 is provided. The accurate positions of the manufacturing and installation of each component are obtained through the calibration process, eliminating the influence of installation and manufacturing errors on the alignment accuracy and improving the alignment accuracy.

[0103] The following uses the first reference hole and the second reference hole to represent the two reference holes 231 provided on the detection jig 23 respectively.

[0104] Specifically, as Figure 3 shown, in step S2: According to the offset of one of the reference holes 231 on the detection jig 23 relative to the center of the first camera's field of view before and after the rotation of the rotary table 22 and the coordinates of the second camera 21, calculate the position offset of the moving platform relative to the second camera 21 when the rotary table 22 is in the initial position. Specifically, it includes:

[0105] S21: The rotary table 22 returns to the reference position, and the X-axis of the gantry 1 and the Y11-axis of the detection unit group 2 return to the reference points again;

[0106] S22: Move the X-axis of the gantry 1 and the Y11-axis of the detection unit group 2 so that the first reference hole at the lower end of the detection jig 23 appears in the field of view of the first camera; record the offset of the first reference hole relative to the center of the first camera's field of view at this time, denoted as (h1x, h1y); record the coordinates of the X-axis of the gantry 1 and the Y11-axis of the detection unit group 2 at this time, denoted as b1x, b1y;

[0107] S23: Calculate the offset of the first reference hole relative to the second camera 21 as (off1x, off1y) according to (Aoffsx, Aoffsy), (h1x, h1y), b1x, b1y, where:

[0108] off1x = h1x + Aoffsx - b1x;

[0109] off1y = h1y + Aoffsy - b1y;

[0110] S24: Manually operate the moving platform of the rotary table 22 to rotate by θ angle, and this θ angle is a definite value, and the counterclockwise direction is specified as the positive direction;

[0111] S25: Move the X-axis of the gantry 1 and the Y11-axis of the detection unit group 2 to make the first reference hole 231 at the lower end of the detection jig 23 appear in the field of view of the first camera again; record the offset of the first reference hole relative to the center of the first camera's field of view at this time, denoted as (h2x, h2y); record the coordinates of the X-axis of the gantry 1 and the Y11-axis of the detection unit group 2 at this time, denoted as b2x, b2y;

[0112] S26: Calculate the offset of the first reference hole relative to the second camera 21 as (off2x, off2y) according to (Aoffsx, Aoffsy), (h2x, h2y), b2x, b2y, where:

[0113] off2x = h2x + Aoffsx - b2x;

[0114] off2y = h2y + Aoffsy - b2y;

[0115] S27: Calculate the position offset (X UvwCam , Y UvwCam ) of the moving platform relative to the second camera 21 when the rotary table 22 is in the initial position based on (off1x, off1y), (off2x, off2y), and θ, where:

[0116] X UvwCam = avgx - Δy / (2 × tan(θ / 2));

[0117] Y UvwCam = avgy + Δx / (2 × tan(θ / 2));

[0118] Δx = off2x - off1x;

[0119] Δy = off2y - off1y;

[0120] avgx = (off1x + off2x) / 2;

[0121] avgy = (off1y + off2y) / 2.

[0122] According to steps S21 to S27, observe the position change of the first reference hole on the detection jig 23 when the rotary table 22 is in different positions through the first camera, and calibrate the position of the moving platform of the rotary table 22 relative to the second camera 21 in its initial state, providing a basis for the position calibration of the detection jig 23.

[0123] Specifically, as Figure 4 shown, in step S3: Calculate the angle between the X-axis direction of the rotary table 22 and the X-axis direction of the device coordinate system according to the offset of one of the reference holes 231 on the detection jig 23 relative to the center of the field of view of the first camera before and after the rotary table 22 moves along the X-axis. Specifically, it includes:

[0124] S31: Manually operate the rotary table 22 to return to the reference position, and manually operate the moving platform of the rotary table 22 to move a certain distance in the positive X direction relative to the fixed platform of the rotary table 22;

[0125] S32: Move the X-axis of the gantry 1 and the Y11-axis of the detection unit group 2 to make the first reference hole at the lower end of the detection jig 23 appear in the field of view of the first camera again; Record the offset of the first reference hole relative to the center of the field of view of the first camera at this time, denoted as (h3x, h3y); Record the coordinates of the X-axis of the gantry 1 and the Y11-axis of the detection unit group 2 at this time, denoted as b3x, b3y;

[0126] S33: Calculate the offset of the first reference hole relative to the second camera 21 as (off3x, off3y) based on (Aoffsx, Aoffsy), (h3x, h3y), b3x, and b3y, where:

[0127] off3x = h3x + Aoffsx - b3x;

[0128] off3y = h3y + Aoffsy - b3y;

[0129] S34: Calculate the angle Th between the X-axis direction of the rotary table 22 itself and the X-axis direction of the device coordinate system based on (off1x, off1y) and (off3x, off3y) UvwCam , where:

[0130] Th UvwCam = atan2((off3y - off1y), (off3x - off1x)).

[0131] where atan2 is the four-quadrant arctangent function.

[0132] According to steps S31 to S34, by observing the position change of the first reference hole on the detection jig 23 when the rotary table 22 is in different positions through the first camera, calibrate the angle between the X-axis direction of the rotary table 22 itself and the X-axis direction of the device coordinate system, providing a basis for the angle calibration of the detection jig 23.

[0133] Specifically, in step S4: According to the offsets of the two reference holes 231 relative to the center of the field of view of the first camera observed by the first camera, calculate the pose of the detection jig 23 relative to the rotary table 22 when the rotary table 22 is in the initial position. Since there are multiple probe needles 232 on the detection jig 23, to uniformly represent the pose of the entire detection jig 23, such as Figure 5 Establish a jig coordinate system. Take the midpoint of the two reference holes 231 as the origin and the connection line of the two reference holes 231 as the X-axis. Use the X j O j Y j pose of the jig coordinate system relative to the rotary table 22 to represent the pose of the detection jig 23 relative to the rotary table 22; step S4 specifically includes steps S41 to S44:

[0134] S41: The rotary table 22 returns to the reference position, and operate the X-axis of the gantry 1 and the Y11 of the detection unit group so that the second reference hole at the lower end of the jig appears in the field of view of the first camera; record the offset of the second reference hole relative to the center of the field of view of the first camera at this time, denoted as (h4x, h4y); record the coordinates of the X-axis of the gantry 1 and the Y11 axis of the detection unit group 2 at this time, denoted as b4x, b4y;

[0135] S42: Calculate the offset of the second reference hole relative to the second camera 21 as (off4x, off4y) based on (Aoffsx, Aoffsy), (h4x, h4y), b4x, and b4y, where:

[0136] off4x = h4x + Aoffsx - b4x;

[0137] off4y = h4y + Aoffsy - b4y;

[0138] S43: Calculate the offset of the detection jig 23 relative to the rotary table 22 as (X UvwCam , Y UvwCam ) based on (off1x, off1y), (off4x, off4y), (X JigUvw , Y JigUvw ), where:

[0139] X JigUvw = off1x / 2 + off4x / 2 - X UvwCam ;

[0140] Y JigUvw = off1y / 2 + off4y / 2 - Y UvwCam ;

[0141] S44: Calculate the angle TH UvwCam that the detection jig 23 rotates relative to the rotary table 22 based on (off1x, off1y), (off4x, off4y), and Th JigUvw , where:

[0142] TH JigUvw = atan2((off4y - off1y), (off4x - off1x)) - Th UvwCam ;

[0143] where atan2 is the four - quadrant arctangent function.

[0144] Through the above steps S1 to S4, calibrate the detection unit group 2. When there are two or more detection unit groups 2, repeat steps S1 to S4 to perform the above calibration on other detection unit groups 2, so that during the operation of the equipment, the poses of the detection unit group 2 and the rotary table 22 can be inversely obtained in combination with the above calibration results when the detection jig 23 is aligned with the detection test area, realizing the simultaneous and accurate alignment of multiple probe needles 232 to the test area. This ensures the alignment accuracy between the detection unit group 2 and the product to be detected during the detection process.

[0145] Therefore, when using the detection equipment calibrated by the above calibration method for detection, the method for accurately aligning its multiple probe needles 232 with the product to be detected is as follows:

[0146] A method for aligning a display panel detection device comprises the following steps:

[0147] S100: Observe multiple GMark coordinates of the loaded substrate 3 through the second camera 21 to determine the coordinates of multiple test points in a test area;

[0148] The second camera 21 is used to observe multiple GMark coordinates of the substrate 3 to be tested after loading, and multiple test points of a test area are determined according to the GMark coordinates and the design drawing of the substrate 3 to be tested, for example Figure 6 The coordinates of T1 and T2 in the test area; or directly determine the coordinates of multiple test points such as T1 and T2 in a test area through observation by the second camera 21. Then determine the test area coordinate system X based on the coordinates of multiple test points according to the following principles C O C Y C 's position (X JigDev ,Y JigDev , TH JigDev ), which is also the target posture of the detection fixture 23.

[0149] S200: Determine the test area coordinate system X according to the coordinates of multiple test points C O C Y C 's position (X JigDev ,Y JigDev , TH JigDev ), the test area coordinate system pose (X JigDev ,Y JigDev , TH JigDev ) as the target posture of the detection fixture 23;

[0150] The test point and test area coordinate system X are set C O C Y C The relative position relationship between the test point and the coordinate system X of the probe 232 and the detection fixture 23 j O j Y j The relative position relationship between is consistent, such as Figure 6 As shown; it should be noted that there are many test points in each test area. There is a fixed relative position relationship between these test points. When describing the position of the test area, it would be cumbersome and unnecessary to describe the position of each test point. The position and posture of the entire area can be described as a whole. For this reason, it is assumed that a coordinate system is established near the test area according to certain rules, so that all the measuring points in the test area have a fixed relative position relationship with the coordinate system. At this time, the position and posture of the coordinate system can uniquely determine the position of all the measuring points in the test area. In other words, this coordinate system represents the "position and posture of all test points." Figure 6 ​Two test area coordinate systems are illustrated in the figure. It can be seen therefrom that there is a fixed relative position relationship between such a coordinate system and all the test points of the test area it represents, which is denoted as R1 here. The detection fixture 23 has a lot of probe needles 232, and there is also a relative position relationship between the probe needles 232 and the fixture coordinate system XjOjYj, which is denoted as R2. Obviously, if R1 and R2 are the same, then by moving the detection fixture 23 to make the coordinate system XjOjYj coincide with the test area coordinate system, all the probe needles 232 on the detection fixture 23 will align with all the test points of the test area. That is to say, the test area coordinate system is calculated based on the position relationship R2 and the coordinates of multiple test points.

[0151] Therefore, the relative position relationship between the test points and the test area coordinate system X C O C Y C is consistent with the relative position relationship between the corresponding probe needles 232 of the test points and the detection fixture 23 coordinate system X j O j Y j In this way, when moving the gantry 1 and rotating the rotary table 22 to make the pose of the detection fixture 23 coordinate system the same as the pose of the test area coordinate system, each probe needle 232 of the detection fixture 23 can align with each test point simultaneously.

[0152] S300: Calculate the X-axis coordinate of the gantry, the Y11-axis coordinate of the detection unit group, and the pose of the rotary table according to the above calibration method and the target pose of the detection fixture 23.

[0153] During the operation of the equipment, the pose of the detection fixture 23 in the equipment can be determined according to the rotation axis of the rotary table 22 itself, its own X-axis, the X-axis of the gantry 1, and the Y11-axis coordinate of the detection unit group 2. At the same time, when the poses of multiple test points in the test area are known, the coordinates of each axis of the gantry 1 and the rotary table 22 can also be calculated. The equipment moves according to this coordinate to make the pose of the detection fixture 23 consistent with the poses of multiple test points, realizing simultaneous alignment of the two. At this time, the target pose of the detection fixture 23, that is, the poses of multiple test points, is: (X JigDev , Y JigDev , TH JigDev ).

[0154] There are the following two methods for obtaining the X-axis of the gantry 1, the Y11-axis of the detection unit group 2, and the pose of the rotary table 22 according to the poses of multiple test points during the operation of the equipment:

[0155] The first method: Assume that the rotary table 22 only rotates. From the offset (X JigUvw , Y JigUvw ) of the detection fixture (23) relative to the rotary table (22) and the target pose (X JigDev , YJigDev , TH JigDev ), and the pose (X UvwCam , Y UvwCam , TH UvwCam ) of the moving platform relative to the second camera (21) when the rotary table (22) is in the initial position to inverse solve the X-axis coordinate of the gantry (1), the Y11-axis coordinate of the detection unit group (2), and the rotation axis coordinate of the rotary table (22):

[0156] θ = TH JigDev -TH JigUvw -TH UvwCam

[0157] X JigCam = X JigUvw cosθ - Y JigUvw sinθ + X UvwCam

[0158] Y JigCam = X JigUvw sinθ + Y JigUvw cosθ + Y UvwCam

[0159] Y11 = Y JigDev -Y JigCam

[0160] X = X JigDev -X JigCam

[0161] where θ is the rotation angle of the rotary table 22, X represents the X coordinate of the X-axis of the gantry 1, Y11 represents the Y coordinate of the Y11-axis of the detection unit group 2, and (X JigCam , Y JigCam ) represents the offset of the detection jig 23 relative to the second camera 21 in the device coordinate system.

[0162] The second method: Assume that the X-axis of the gantry 1 is determined. From the offset and rotation angle (X JigUvw , Y JigUvw , TH JigUvw ) of the detection jig 23 relative to the rotary table 22, the offset and rotation angle (X UvwCam , Y UvwCam , TH UvwCam ) of the rotary table 22 relative to the second camera 21, and the target pose (X JigDev , Y JigDev , TH JigDev ) of the detection jig 23 to inverse solve the pose (X Uvw , θ) of the rotary table 22 and the Y11-axis coordinate of the detection unit group 2:

[0163] θ = TH JigDev -THJigUvw -TH UvwCam

[0164] X JigDev = X + X JigCam

[0165] X JigCam = X JigUvw cosθ - Y JigUvw sinθ + X Uvw cosTH UvwCam + X UvwCam

[0166] Y JigCam = X JigUvw sinθ + Y JigUvw cosθ + X Uvw sinTH UvwCam + Y UvwCam

[0167] Y11 = Y JigDev - Y JigCam

[0168] where θ is the rotation angle of the rotary table 22, X represents the X coordinate of the X - axis of the gantry 1, X Uvw represents the movement target in the X - direction of the rotary table 22 itself, and Y11 represents the Y coordinate of the Y11 - axis of the detection unit group 2.

[0169] According to the above two methods, during the operation of the equipment, the X - axis of the gantry 1, the Y11 - axis of the detection unit group 2 and the rotation axis coordinates of the rotary table are obtained based on multiple test point postures, so as to achieve the accurate alignment of the probe 232 on the detection jig 23 with the test area by controlling the movement of the gantry 1, the detection unit group 2 and the movement and rotation of the rotary table 22.

[0170] During the equipment detection process, A) for the gantry 1 with only one detection unit group 2, the gantry 1 obtains the set position according to the equilibrium calculation, and the X - axis of the gantry 1 moves to this set position. Before the detection jig 23 moves to the target point, first calculate (X Uvw , θ) and the Y11 - axis coordinate of the detection unit group 2 according to the second method. If X Uvw does not exceed the travel range of the rotary table 22, then the rotary table 22 and the detection unit group 2 move according to the calculated coordinates. At this time, the gantry 1 does not need to translate, which has the characteristic of rapid positioning. If X UvwIf it exceeds the travel range of the rotary table 22, the rotation angle θ of the rotary table 22 and the coordinate of the X-axis of the gantry 1 are calculated according to the first method, and the coordinate of the Y11-axis of the detection unit group 2 is detected. The moving platform of the rotary table 22 rotates by an angle θ, the X-axis of the rotary table 22 stops at the initial position, the X-axis of the gantry 1 moves to the X position, and the Y11-axis of the detection unit group 2 moves to the Y11 position, so that multiple probe needles 232 on the detection jig 23 can be aligned with multiple test points.

[0171] If there are multiple test areas with approximate X coordinates on the substrate to be tested, first calculate multiple X coordinate values of the gantry 1 according to the first method. Perform an equalization calculation on the multiple X coordinate values, and take the midpoint of the multiple X coordinate values as the target position of the gantry 1 to move the gantry. Then calculate the poses (X Uvw , θ) of the rotary table 22 corresponding to each test area and the coordinate of the Y11-axis of the detection unit group 2 in turn according to the second method, and move the rotary table 22 and the Y11-axis of the detection unit group 2 to the corresponding positions, so that the detection jig 23 can be aligned with multiple test areas in turn. Realize moving the gantry 1 once and detecting multiple test areas.

[0172] B) For a device with two detection unit groups 2 on the same gantry 1, the gantry 1 can be moved once to complete the detection of multiple test areas with approximate X coordinates. Because their X coordinates are approximate, these test areas are hereinafter referred to as "a row". During this period, the X coordinate of the X-axis of the gantry 1 does not change. When the X coordinate of the detection jig 23 needs to be adjusted, it is achieved by the movement of the X-axis of the rotary table 22. Only when the travel of the X-axis is not enough to align the detection jig 23 with the test point, the gantry 1 is moved.

[0173] To achieve this, before the gantry 1 moves towards a row, first perform an equalization calculation of the X coordinate of the gantry 1 to obtain the set position, the X-axis of the gantry 1 moves to this set position, and then use the second method to find the θ, X Uvw and Y11 coordinates when detecting each detection area, and cooperate with adjusting the poses (X Uvw , θ) of the rotary table 22 of each detection unit group 2 and the Y11-axis of each detection unit group 2 on the gantry 1 to achieve the purpose of completing the testing of multiple test areas by moving the X-axis of the gantry 1 once; when X Uvw exceeds the travel range of the rotary table 22, then calculate the rotation angle θ of the rotary table 22 and the coordinate of the X-axis of the gantry 1, and the coordinate of the Y11-axis of the detection unit group 2 according to the first method. The moving platform of the rotary table 22 rotates by an angle θ, the X-axis of the rotary table 22 stops at the initial position, the X-axis of the gantry 1 moves to the X position, and the Y11-axis of the detection unit group 2 moves to the Y11 position, so that multiple probe needles 232 on the detection jig 23 can be aligned with multiple test points.

[0174] Specific equilibrium calculation process for the X-axis coordinate of the gantry 1: Assume that the test fixture 23 of the first detection unit 2 needs to detect the test areas A, B, and C successively, and the test fixture 23 of the second detection unit 2 needs to detect the test areas D and E. According to the poses of A, B, and C (the poses of multiple test points in the test area are known) and the calibration results of the first detection unit 2, use the first method to obtain the X coordinates X1, X2, and X3 of the X-axis of the gantry 1; according to the poses of D and E (the poses of multiple test points in the test area are known) and the calibration results of the second detection unit 2, use the first method to obtain the X coordinates X4 and X5 of the X-axis of the gantry 1. If the difference between MAX(X1, X2, X3, X4, X5) and MIN(X1, X2, X3, X4, X5) is less than the X-direction movement stroke limit of the rotary table 22, the X-axis of the gantry 1 can be moved to (MAX(X1, X2, X3, X4, X5) + MIN(X1, X2, X3, X4, X5)) / 2. This position is the set position where the X-axis of the gantry 1 first moves to for a device with two or more detection units 2 on one gantry 1.

[0175] It should be noted that for a device with two gantries 1, calculate the target coordinates according to the detection needs respectively, and move the two gantries 1 for detection respectively. The target position of each gantry 1 movement is calculated according to the number of test units on the gantry 1, using the alignment method for a gantry 1 with only one detection unit 2 or for a gantry 1 with two detection units 2 on the same gantry 1.

[0176] In this embodiment, a test fixture 23 has multiple probing needles 232, which improves the detection efficiency. By adding a rotary table 22, the test fixture 23 realizes the X-direction movement and the rotation around the Z-axis of the moving platform of the rotary table 22, enabling the multiple probing needles 232 of the test fixture 23 to be accurately aligned with multiple test points in the test area simultaneously. In addition, a calibration method for the installation positions of the rotary table 22 and the test fixture 23 is provided. The accurate manufacturing and installation positions of each component are obtained through the calibration process, eliminating the influence of installation and manufacturing errors on the alignment accuracy and improving the alignment accuracy. During the device detection process, using coordinate equilibrium calculation reduces the overall movement of the gantry 1 in the X direction, shortening the positioning time of the detection process.

[0177] In this embodiment, as an alternative to the detection jig 23: No reference holes 231 are provided on the detection jig 23. During the calibration process, a special marking pattern is temporarily fixed on the detection jig 23. Compared with steps S2 and S3, in this solution, by observing the position change of the special marking pattern after the movement of the rotary table 22, the position of the moving platform of the rotary table 22 relative to the second camera 21 and the X-axis direction in the initial state are calibrated. At the same time, compared with step S4, here the pose of the jig relative to the rotary table 22 is determined by observing the positions of multiple probe needles 232 with the first camera. However, the results of observing the probe needles 232 in this solution are easily affected by light, the shape of the needles, and the field of view background. The process is complex and the accuracy is relatively low. But it does not require the manufacturing of reference holes 231 on the detection jig 23. Therefore, in the actual production process, it is possible to choose whether to set reference holes 231 on the detection jig 23 for equipment calibration and alignment according to needs.

[0178] In addition, in this embodiment, as an alternative to the rotary table 22: For the gantry 1 with only one detection unit group 2, the rotary table 22 may not include the movement in the X direction, and the structure is simpler. However, at this time, when the equipment moves from one test area to another test area. Even if the X coordinates of the two test areas are very close, the entire gantry 1 needs to move. The gantry 1 is heavy, and it takes a longer stabilization time to move to the new position. Moreover, this detection unit group 2 without X-direction movement is not applicable to the case where there are more than two detection heads on one gantry 1. Since the substrate to be detected is more or less inclined, if there are two such detection unit groups 2 on the same gantry 1, it will be impossible to align the two jigs with the test area at the same time, impossible to perform detection simultaneously, and impossible to cooperate for detection. Therefore, in the actual production process, it is possible to choose whether to set the movement in the X direction on the rotary table 22 for equipment calibration and alignment according to needs. It should be noted that the rotary table 22 set in this embodiment includes the movement in the X direction, and this X-axis direction can be used in cooperation with the X-axis direction of the gantry 1. When the test areas are relatively close, it is possible to preferentially move the X direction on the rotary table 22 to move the detection jig 23 to the next test area, reducing the movement of the X axis of the gantry 1 and improving the detection efficiency. Briefly understood, the X direction on the rotary table 22 can be used as a fine adjustment of the X-axis movement of the gantry 1 to improve the alignment accuracy and efficiency of the entire detection equipment.

[0179] In addition, it should be noted that the installation base of the detection unit group 2 in this embodiment can be a structure different from that of the gantry 1. As long as the detection unit group 2 can move in the X and Y directions.

[0180] As described above, it is only the preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes, shall be covered by the protection scope of the present invention.

Claims

1. A calibration method for an alignment device of a display panel detection device, characterized in that: The alignment device comprises a first camera and a gantry (1) movable along the X-direction, the gantry (1) being provided with one or more detection unit groups (2), each detection unit group (2) comprising a rotating worktable (22) movable along the Y-axis and capable of lifting and lowering, and a second camera (21), a detection fixture (23) being provided on the movable platform of the rotating worktable (22), the movable platform of the rotating worktable (22) being movable along the X-direction and also capable of rotating about the Z-axis, and two reference holes (231) and two or more measuring pins (232) being provided on the detection fixture (23); The calibration method of the alignment device comprises the following steps: S1: calibrating the installation position of the first camera and the origin position of the second camera (21) by using a marking pattern C; S2: Calculate the position offset of the moving platform relative to the second camera (21) when the rotating table (22) is in the initial position based on the offset of one of the reference holes (231) on the inspection fixture (23) observed by the first camera relative to the center of the field of view of the first camera before and after the rotating table (22) rotates and the coordinates of the second camera (21); S3: Calculate the angle between the X-axis direction of the rotating table (22) and the X-axis direction of the device coordinate system based on the offset of one of the reference holes (231) on the inspection fixture (23) observed by the first camera relative to the center of the first camera's field of view when the rotating table (22) moves forward and backward along the X-axis and the coordinates of the second camera (21); S4: Calculate the position of the detection fixture (23) relative to the rotating table (22) when the rotating table (22) is in the initial position based on the offset of the two reference holes (231) observed by the first camera relative to the center of the first camera's field of view and the coordinates of the second camera (21); S5: Repeat steps S1 to S4 to calculate the positions of other inspection jigs (23) relative to the rotating worktable (22) and the positions of the rotating worktable (22) relative to the second camera (21).

2. The calibration method according to claim 1, characterized in that: In step S1: the installation position of the first camera and the origin position of the second camera (21) are calibrated by using the marking pattern C, wherein the installation position of the first camera is calibrated as follows: The center of the marking pattern C is used as the origin, the movement direction of the X-axis of the gantry (1) is used as the X-axis, and the movement direction of the Y11 axis of the detection unit group (2) is used as the Y-axis to establish a device coordinate system; Adjust the installation position of the first camera so that the center of the field of view of the first camera is close to the marking pattern C; The offset of the center of the field of view of the first camera relative to the marking pattern C is determined and recorded as (Aoffsx, Aoffsy).

3. The calibration method according to claim 1, characterized in that: In step S1: the installation position of the first camera and the origin position of the second camera (21) are calibrated by using the marking pattern C, wherein the calibration of the origin position of the second camera (21) is specifically as follows: The X-axis of the gantry (1) and the Y11 axis of the detection unit (2) return to the reference point, and the offset of the reference point of the X-axis of the gantry (1) relative to the X-axis of the equipment coordinate system is recorded as X RefOffs0 , record the offset of the reference point of the Y11 axis of the detection unit group (2) relative to the Y axis of the device coordinate system as Y11 RefOffs0 ; The detection unit group (2) is moved to align the center of the field of view of the second camera (21) with the marking pattern C, the origin position of the second camera (21) is calibrated, and the coordinate of the X axis of the gantry (1) at this time is recorded as x a , the Y11 axis coordinate of the detection unit group (2) is y b ; Modify the offset of the reference point of the X-axis of the gantry (1) relative to the X-axis of the device coordinate system to (X RefOffs0 -x a ), modify the offset of the reference point of the Y11 axis of the detection unit group (2) relative to the Y axis of the device coordinate system to (Y11 RefOffs0 -y b ), coordinate (X RefOffs0 -x a ,Y11 RefOffs0 -y b ) is used as the reference point position after calibration of the second camera (21).

4. The calibration method according to claim 2, characterized in that: In step S2: based on the offset of one of the reference holes (231) on the inspection fixture (23) observed by the first camera relative to the center of the field of view of the first camera before and after the rotation of the rotary table (22) and the coordinates of the second camera (21), the position offset of the moving platform relative to the second camera (21) when the rotary table (22) is in the initial position is calculated, specifically including: The rotary table (22) returns to the reference position, and the X-axis of the gantry (1) and the Y11-axis of the detection unit group (2) return to the reference point; Moving the X-axis of the gantry (1) and the Y11-axis of the detection unit group (2) so that the reference hole (231) at the lower end of the detection fixture (23) appears in the field of view of the first camera; Record the offset of the reference hole (231) relative to the center of the first camera field of view at this time, set to (h1x, h1y); Record the coordinates of the X-axis of the gantry (1) and the Y11-axis of the detection unit group (2) at this time, and set them as b1x and b1y; The offset of the reference hole (231) relative to the second camera (21) is calculated as (off1x, off1y) according to (Aoffsx, Aoffsy), (h1x, h1y), b1x, b1y, wherein off1x=h1x+Aoffsx-b1x, off1y=h1y+Aoffsy-b1y; The movable platform of the rotary table (22) is operated to rotate through an angle θ; Moving the X-axis of the gantry (1) and the Y11-axis of the detection unit group (2) so that the same reference hole (231) at the lower end of the detection fixture (23) appears in the field of view of the first camera again; Record the offset of the reference hole (231) relative to the center of the first camera field of view at this time, set to (h2x, h2y); Record the coordinates of the X-axis of the gantry (1) and the Y11-axis of the detection unit group (2) at this time, and set them as b2x and b2y; The offset of the reference hole (231) relative to the second camera (21) is calculated as (off2x, off2y) according to (Aoffsx, Aoffsy), (h2x, h2y), b2x, and b2y, wherein off2x = h2x + Aoffsx - b2x, and off2y = h2y + Aoffsy - b2y; The position offset (X) of the moving platform relative to the second camera (21) when the rotating table (22) is in the initial position is calculated based on (off1x, off1y), (off2x, off2y) and θ. UvwCam , Y UvwCam ), where X UvwCam = avgx-Δy / (2×tan(θ / 2), Y UvwCam = avgy+Δx / (2×tan(θ / 2)), Δx=off2x-off1x, Δy=off2y-off1y, avgx=(off1x+off2x) / 2, avgy=(off1y+off2y) / 2.

5. The calibration method according to claim 4, characterized in that: In step S3: based on the offset of one of the reference holes (231) on the inspection fixture (23) observed by the first camera relative to the center of the field of view of the first camera when the rotating table (22) moves forward and backward along the X-axis and the coordinates of the second camera (21), the angle between the X-axis direction of the rotating table (22) and the X-axis direction of the device coordinate system is calculated, specifically including: The rotary table (22) returns to the reference position, and the moving platform of the rotary table (22) is operated to move a certain distance in the positive X direction relative to the fixed platform of the rotary table (22); Moving the X-axis of the gantry (1) and the Y11-axis of the detection unit group (2) so that the same reference hole (231) at the lower end of the detection fixture (23) appears in the field of view of the first camera; Record the offset of the reference hole (231) relative to the center of the first camera field of view at this time, set to (h3x, h3y); Record the coordinates of the X-axis of the gantry (1) and the Y11-axis of the detection unit group (2) at this time, and set them as b3x and b3y; The offset of the reference hole (231) relative to the second camera (21) is calculated as (off3x, off3y) according to (Aoffsx, Aoffsy), (h3x, h3y), b3x, and b3y, wherein off3x=h3x+Aoffsx-b3x, and off3y=h3y+Aoffsy-b3y; The angle Th between the X-axis direction of the rotary table (22) and the X-axis direction of the device coordinate system is calculated based on (off1x, off1y) and (off3x, off3y). UvwCam , where Th UvwCam = atan2((off3y-off1y), (off3x-off1x)).

6. The calibration method according to claim 1, characterized in that: In step S4: based on the offset of the two reference holes (231) observed by the first camera relative to the center of the field of view of the first camera and the coordinates of the second camera (21), the position of the detection fixture (23) relative to the rotating table (22) when the rotating table (22) is in the initial position is calculated, which specifically includes: The rotating worktable (22) returns to the reference position, and the X-axis of the gantry (1) and the detection unit group (2) Y11 are operated so that another reference hole (231) at the lower end of the fixture appears in the field of view of the first camera; Record the offset of the reference hole (231) relative to the center of the first camera field of view at this time, set to (h4x, h4y); Record the coordinates of the X-axis of the gantry (1) and the Y11-axis of the detection unit group (2) at this time, and set them as b4x and b4y; The offset of the reference hole (231) relative to the second camera (21) is calculated as (off4x, off4y) according to (Aoffsx, Aoffsy), (h4x, h4y), b4x, and b4y, wherein off4x=h4x+Aoffsx-b4x, and off4y=h4y+Aoffsy-b4y; According to (off1x, off1y), (off4x, off4y), (X UvwCam , Y UvwCam ) calculates the offset (X) of the inspection fixture (23) relative to the rotary table (22) JigUvw , Y JigUvw ), where X JigUvw = off1x / 2+off4x / 2-X UvwCam , Y JigUvw =off1y / 2+off4y / 2-Y UvwCam ; According to (off1x, off1y), (off4x, off4y), Th UvwCam The angle TH that the inspection fixture (23) rotates relative to the rotary table (22) is calculated. JigUvw , where TH JigUvw = atan2((off4y-off1y), (off4x-off1x))-Th UvwCam .

7. The alignment method of the alignment device of the display panel detection equipment according to claim 1, characterized in that: The steps include: Observing multiple GMark coordinates of the loaded substrate (3) through a second camera (21) to determine the coordinates of multiple test points in a test area; Determine the test area coordinate system X based on the coordinates of multiple test points C O C Y C The pose (X JigDev , Y JigDev , TH JigDev ), the test area coordinate system pose (X JigDev , Y JigDev , TH JigDev ) as the target posture of the detection fixture (23); According to the calibration method of claim 2 and the target posture of the detection fixture (23), the gantry X-axis coordinates, the detection unit group Y11-axis coordinates and the posture of the rotary table are calculated.

8. The alignment method of the display panel detection device according to claim 7, characterized in that: In the calibration method according to claim 2 and the target posture calculation of the detection fixture (23) for calculating the gantry X-axis coordinate, the detection unit group Y11-axis coordinate and the posture of the rotary table, the method specifically includes: Assuming that the X-axis of the gantry (1) is fixed, the offset and rotation angle (X JigUvw , Y JigUvw , TH JigUvw ), the offset and rotation angle (X) of the rotating table (22) relative to the second camera (21) UvwCam , Y UvwCam , TH UvwCam ) and the target position (X JigDev , Y JigDev , TH JigDev ), reverse the position (X) of the rotary table (22) Uvw ,θ) and the Y11 axis coordinate of the detection unit group (2).

9. The alignment method of the display panel detection device according to claim 7, characterized in that: In the calibration method according to claim 2 and the target posture calculation of the detection fixture (23) for calculating the gantry X-axis coordinate, the detection unit group Y11-axis coordinate and the posture of the rotary table, the method specifically includes: Assuming that the rotary table (22) only rotates, the offset (X JigUvw , Y JigUvw ), the target position (X JigDev , Y JigDev , TH JigDev ) and the position (X) of the moving platform relative to the second camera (21) when the rotating worktable (22) is in the initial position. UvwCam , Y UvwCam , TH UvwCam ) reversely calculate the X-axis coordinate of the gantry (1), the Y11-axis coordinate of the detection unit group (2), and the rotation axis coordinate of the rotary table (22).

10. The alignment method of the display panel detection device according to claim 8, characterized in that: Reverse the position of the rotary table (22) (X Uvw ,θ) and the specific formula of the Y11 axis coordinate of the detection group (2) are as follows: θ=TH JigDev -TH JigUvw -TH UvwCam Wherein, θ is the rotation angle of the rotary table (22), X represents the X coordinate of the X axis of the gantry (1), and X Uvw represents the X-direction movement target of the rotating worktable (22) itself, and Y11 represents the Y coordinate of the Y11 axis of the detection unit group (2).

11. The alignment method of the display panel detection device according to claim 9, characterized in that: The specific formulas for inversely calculating the X-axis coordinates of the gantry (1), the Y11-axis coordinates of the detection unit (2), and the rotation axis coordinates of the rotary table (22) are as follows: θ=TH JigDev -TH JigUvw -TH UvwCam Wherein, θ is the rotation angle of the rotary table (22), X represents the X coordinate of the X axis of the gantry (1), and Y11 represents the Y coordinate of the Y11 axis of the detection unit group (2). JigCam , Y JigCam ) represents the offset of the detection fixture (23) relative to the second camera (21) in the device coordinate system.

12. The alignment method of the display panel detection device according to claim 9, characterized in that: For multiple test areas with similar X coordinates on the substrate to be tested, according to the method of inversely calculating the X-axis coordinate of the gantry (1), the Y11-axis coordinate of the detection unit group (2), and the rotation axis coordinate of the rotary table (22) as described in claim 9, multiple X-coordinate values ​​of the gantry (1) are calculated, and the target position of the gantry (1) is calculated by balancing the multiple X-coordinate values, and then the position (X) of the rotary table (22) is inversely calculated as described in claim 8. Uvw ,θ) and the Y11-axis coordinate of the detection unit group (2), and calculate the position and posture (X Uvw ,θ) and the Y11 axis coordinate of the detection unit group (2); Move the gantry (1) to the target position of the gantry (1), and then move the rotary table (22) and the detection unit group to the position (X Uvw ,θ) and the Y11-axis coordinate of the detection unit group (2), so that the detection fixture (23) is aligned with multiple test areas in sequence.

Images