Bent needle point calibration equipment and calibration method based on vision

Abstract

The invention relates to the technical field of equipment calibration, and discloses vision-based curved needle point calibration equipment and a calibration method. The vision-based bent needle point calibration equipment comprises an automatic mechanism device, a glue barrel bent needle point device and a processing device, and the glue barrel bent needle point device is hung on the automatic mechanism device and comprises a bent needle point; the processing device is used for sending a control instruction to the automatic mechanism device; the automatic mechanism device is used for moving according to the control instruction, collecting a bent needle point image of the bent needle point and transmitting the bent needle point image to the processing device; and the processing device is also used for processing the bent needle point image based on a preset image processing algorithm, identifying the bent needle point position in the bent needle point image, and calibrating the bent needle point based on the bent needle point position in the bent needle point image. Therefore, effective and accurate equipment calibration is carried out by accurately identifying and grabbing the position of the bent needle point by utilizing a mechanical structure in cooperation with an industrial vision technology, the dispensing precision and accuracy of the dispensing equipment can be improved, and the product quality is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment calibration, and in particular to a vision-based curved needle tip calibration device and a calibration method. Background Art

[0002] In the dispensing equipment assembly needle valve, the curved needle tip is a component of the assembly needle valve. The glue discharge of the curved needle tip plays a key role in the precise dispensing of the side or gap of the product. However, due to the special shape and angle of the curved needle tip, if the traditional straight curved needle tip calibration method is used to calibrate the curved needle tip, it will result in slow calibration speed, low efficiency, more human intervention, and strong uncontrollability, which cannot provide more convenient and accurate calibration operations for precise dispensing. Summary of the invention

[0003] The embodiments of the present invention aim to provide a vision-based curved needle tip calibration device and calibration method, which can solve the problems of slow calibration speed, low efficiency, frequent human intervention, and inability to provide more convenient and accurate correction operations for precision dispensing by using the traditional straight curved needle tip calibration method to calibrate the curved needle tip.

[0004] In order to solve the above technical problems, the first embodiment of the present invention provides a vision-based curved needle tip calibration device, including an automated mechanism device, a rubber barrel curved needle tip device and a processing device, wherein:

[0005] The glue barrel curved needle tip device is mounted on the automated mechanism device and includes a curved needle tip;

[0006] The processing device is communicatively connected with the automation mechanism device, and is used to send control instructions to the automation mechanism device;

[0007] The automated mechanism device is used to move according to the control instruction, collect the curved needle tip image of the curved needle tip, and transmit it to the processing device;

[0008] The processing device is also used to perform image processing on the curved needle tip image based on a preset image processing algorithm, identify the curved needle tip position in the curved needle tip image, and calibrate the curved needle tip based on the curved needle tip position in the curved needle tip image.

[0009] Optionally, the automated mechanism device comprises a mobile platform and a camera device, wherein:

[0010] The mobile platform is located on the upper part of the camera device and is used to move the rotation center of the rotation axis to align with the camera center position of the camera device according to the control instruction;

[0011] The photographic device is used to collect the curved needle tip image of the curved needle tip and transmit it to the processing device.

[0012] Optionally, the mobile platform includes an X-axis, a Y-axis, a Z-axis and an R-axis, wherein the effective stroke of the Z-axis is 0 to 360 degrees; the R-axis is a rotation axis with an effective stroke of 0 to 360 degrees, the clockwise direction of the rotation axis is set as the positive direction, and the curved needle tip is rotated to the right direction as the 0 position of the rotation axis.

[0013] Optionally, the photographic device comprises a camera and a coaxial light source, wherein:

[0014] The coaxial light source provides a white coaxial light source for the camera;

[0015] The camera is used to collect the curved needle tip image of the curved needle tip under the illumination of the white coaxial light source provided by the coaxial light source, and transmit it to the processing device.

[0016] Optionally, the glue barrel curved needle tip device further comprises a metal pad and a needle tube, wherein:

[0017] The needle tube is mounted on the Z axis of the mobile platform;

[0018] The curved needle tip is arranged below the needle tube, and the curved needle tip is locked on the needle tube through the metal pad;

[0019] The metal pad is screwed under the needle tube, and an opening is arranged in the middle of the metal pad so that the curved needle tip can pass through the opening and be buckled in the middle of the metal pad.

[0020] Optionally, the back of the metal pad is made of a silver reflective material to ensure that when the camera captures the curved needle tip image of the curved needle tip, the background appears white under the illumination of the white coaxial light source provided by the coaxial light source to reduce noise.

[0021] Accordingly, the second aspect of the present invention provides a vision-based curved needle tip calibration method, which is applied to the vision-based curved needle tip calibration device described in the first aspect of the present invention, and the vision-based curved needle tip calibration device includes an automated mechanism device, a rubber barrel curved needle tip device and a processing device; the vision-based curved needle tip calibration method includes:

[0022] The glue barrel bending needle tip device is mounted on the automated mechanism device, wherein the glue barrel bending needle tip device includes a bending needle tip;

[0023] The processing device sends a control instruction to the automation mechanism device;

[0024] The automated mechanism device moves according to the control instruction, collects the curved needle tip image of the curved needle tip, and transmits it to the processing device;

[0025] The processing device performs image processing on the curved needle tip image based on a preset image processing algorithm, identifies the curved needle tip position in the curved needle tip image, and calibrates the curved needle tip based on the curved needle tip position in the curved needle tip image.

[0026] Optionally, the automated mechanism moves according to the control instruction, collects the curved needle tip image of the curved needle tip, and transmits it to the processing device, including:

[0027] Based on the angle of one rotation of the rotating axis of the mobile platform, the plane on which the rotating axis rotates one circle is evenly divided into a number of rotation positions at a preset rotation angle, wherein the preset rotation angle is the angle of one rotation of the rotating axis of the mobile platform evenly divided by the number of corresponding rotation positions;

[0028] The camera parameters are adjusted, and the image of the curved needle tip at each rotation position is acquired by using the camera and transmitted to the processing device.

[0029] Optionally, the vision-based curved needle tip calibration method also includes the processing device determining a calibration relationship between a camera center and a mobile platform.

[0030] Optionally, the processing device performs image processing on the curved needle tip image based on a preset image processing algorithm, identifies the curved needle tip position in the curved needle tip image, and performs curved needle tip calibration based on the curved needle tip position in the curved needle tip image, including:

[0031] Based on a preset image processing algorithm and a calibration relationship between the camera center and the mobile platform, the curved needle tip is searched from the curved needle tip image at each rotation position to obtain a curved needle tip search result;

[0032] Based on the result of searching for the curved needle tip, the curved needle tip successfully found is calibrated.

[0033] Compared with the prior art, the present invention provides a vision-based curved needle tip calibration device and calibration method. The vision-based curved needle tip calibration device includes an automated mechanism device, a rubber barrel curved needle tip device and a processing device, wherein: the rubber barrel curved needle tip device is mounted on the automated mechanism device, the automated mechanism device moves according to the control instruction of the processing device, and the curved needle tip image of the curved needle tip in the rubber barrel curved needle tip device is collected, the processing device performs image processing on the curved needle tip image based on a preset image processing algorithm, identifies the curved needle tip position in the curved needle tip image, and performs curved needle tip calibration based on the curved needle tip position in the curved needle tip image. Thus, by using the mechanical structure in conjunction with industrial vision technology to accurately identify and grasp the position of the curved needle tip for effective and accurate equipment calibration, it is further possible to provide a stable, reliable, more convenient and accurate correction operation for precise dispensing of product sides or gaps, improve dispensing accuracy, improve the process accuracy and accuracy of the dispensing equipment, and improve product quality; the automated mechanism moves according to the control instructions of the processing device, automatically collects the curved needle tip image of the curved needle tip in the glue barrel curved needle tip device, and the processing device automatically bends the needle tip image based on a preset image processing algorithm. The automated image processing and grasping process reduces manual intervention and improves the working efficiency of the dispensing equipment; the processing device calibrates the curved needle tip based on the curved needle tip position in the curved needle tip image, which can improve the reliability of the system. Thus, it can solve the existing problem of using the traditional straight curved needle tip calibration method to calibrate the curved needle tip, resulting in slow calibration speed, low efficiency, more manual intervention, and failure to provide more convenient and accurate correction operation for precise dispensing. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0035] Figure 1 It is a structural schematic diagram of a curved needle tip calibration device based on vision provided by the present invention;

[0036] Figure 2 It is a schematic flow chart of a curved needle tip calibration method based on vision provided by the present invention;

[0037] Figure 3 It is a 75° curved needle tip image captured by a camera in a curved needle tip calibration method based on vision provided by the present invention;

[0038] Figure 4 It is a 90° curved needle tip image captured by a camera in a curved needle tip calibration method based on vision provided by the present invention.

[0039] The reference numerals are as shown in the following table:

[0040]

[0041] DETAILED DESCRIPTION

[0042] In order to facilitate the understanding of the present invention, the present invention is described in more detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that when an element is described as "fixed to" another element, it can be directly on another element or there can be one or more centered elements therebetween. When an element is described as "connected" to another element, it can be directly connected to another element or there can be one or more centered elements therebetween. The orientation or positional relationship indicated by the terms "upper", "lower", "inner", "outer", "bottom" etc. used in this specification is based on the orientation or positional relationship shown in the accompanying drawings, 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 orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0043] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in this specification and in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used in this specification includes any and all combinations of one or more of the related listed items.

[0044] In addition, the technical features involved in different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0045] In one embodiment, Figure 1 As shown, the present invention provides a vision-based curved needle tip calibration device 100, comprising: an automated mechanism device 10, a rubber barrel curved needle tip device 20 and a processing device 30, wherein:

[0046] The glue barrel bending needle tip device 20 is mounted on the automation mechanism device 10 and includes a bending needle tip 21;

[0047] The processing device 30 is in communication connection with the automation mechanism device 10 and is used to send control instructions to the automation mechanism device 10;

[0048] The automated mechanism device 10 is used to move according to the control instruction of the processing device 30, collect the curved needle tip image of the curved needle tip 21 in the plastic barrel curved needle tip device 20, and transmit it to the processing device 30;

[0049] The processing device 30 is also used to perform image processing on the curved needle tip image based on a preset image processing algorithm, identify the curved needle tip position in the curved needle tip image, and calibrate the curved needle tip based on the curved needle tip position in the curved needle tip image.

[0050] In this embodiment, a vision-based curved needle tip calibration device is provided, including an automated mechanism device, a glue barrel curved needle tip device and a processing device, wherein: the glue barrel curved needle tip device is mounted on the automated mechanism device, the automated mechanism device moves according to the control instruction of the processing device, and the curved needle tip image of the curved needle tip in the glue barrel curved needle tip device is collected, the processing device performs image processing on the curved needle tip image based on a preset image processing algorithm, identifies the curved needle tip position in the curved needle tip image, and performs curved needle tip calibration based on the curved needle tip position in the curved needle tip image. Thus, by using the mechanical structure in conjunction with industrial vision technology to accurately identify and grasp the position of the curved needle tip for effective and accurate equipment calibration, it is further possible to provide a stable, reliable, more convenient and accurate correction operation for precise dispensing of product sides or gaps, improve dispensing accuracy, improve the process accuracy and accuracy of the dispensing equipment, and improve product quality; the automated mechanism moves according to the control instructions of the processing device, automatically collects the curved needle tip image of the curved needle tip in the glue barrel curved needle tip device, and the processing device automatically bends the needle tip image based on a preset image processing algorithm. The automated image processing and grasping process reduces manual intervention and improves the working efficiency of the dispensing equipment; the processing device calibrates the curved needle tip based on the curved needle tip position in the curved needle tip image, which can improve the reliability of the system. Thus, it can solve the existing problem of using the traditional straight curved needle tip calibration method to calibrate the curved needle tip, resulting in slow calibration speed, low efficiency, more manual intervention, and failure to provide more convenient and accurate correction operation for precise dispensing.

[0051] In one embodiment, the automated mechanism device 10 is used to move according to the control instruction of the processing device 30 , collect the curved needle tip image of the curved needle tip 21 in the glue barrel curved needle tip device 20 , and transmit it to the processing device 30 .

[0052] Specifically, Figure 1 As shown, the automated mechanism device 10 includes a mobile platform 11 at the upper part and a camera device 12 at the lower part, wherein:

[0053] The moving platform 11 is located on the upper part of the camera device 12 , and is used to move the rotation center of the rotation axis to align with the camera center position of the camera device 12 according to the control instruction of the processing device 30 .

[0054] Specifically, Figure 1As shown, the mobile platform 11 includes an X-axis, a Y-axis, a Z-axis and an R-axis, wherein the effective stroke of the Z-axis is 0 to 360 degrees. The R-axis is a rotation axis, and the effective stroke is 0 to 360 degrees. The clockwise direction of the rotation axis is set as the positive direction, and the curved needle tip 21 is rotated to the right direction, which is set as the 0 position of the rotation axis.

[0055] The photographing device 12 is used to collect the curved needle tip image of the curved needle tip 21 in the glue barrel curved needle tip device 20 and transmit it to the processing device 30.

[0056] Specifically, Figure 1 As shown, the photographic device 12 includes a camera 121 and a coaxial light source 122 .

[0057] The coaxial light source 122 cooperates with the camera 121 to provide a white coaxial light source for the camera 121 .

[0058] The camera 121 is used to collect the curved needle tip image of the curved needle tip 21 in the glue barrel curved needle tip device 20 under the illumination of the white coaxial light source provided by the coaxial light source 122, and transmit it to the processing device 30.

[0059] For example, camera 121 is a 2D black-and-white camera with 1.3 million pixels fixed vertically downward. In combination with the coaxial light source 122, camera 121 collects the curved needle tip image of the curved needle tip 21 under the illumination of the white coaxial light source provided by the coaxial light source 122 based on a suitable field of view and focal length, and transmits the curved needle tip image to the processing device 30.

[0060] In this embodiment, an automated mechanism device includes a mobile platform on the upper part and a camera device on the lower part, and the mobile platform is mounted with a glue barrel bent needle tip device. The camera collects the bent needle tip image of the bent needle tip in the glue barrel bent needle tip device under the illumination of a white coaxial light source provided by a coaxial light source, thereby realizing automated image capture, reducing manual intervention, and improving the working efficiency of the dispensing equipment.

[0061] In one embodiment, the glue barrel bending needle tip device 20 is mounted on the automated mechanism device 10 .

[0062] Specifically, the glue barrel curved needle tip device 20 is mounted on the Z-axis of the mobile platform, and the glue can be ejected from the curved needle tip 21 along the needle tube under the action of air pressure.

[0063] like Figure 1 As shown, the glue barrel curved needle tip device 20 includes a curved needle tip 21 , a metal pad 22 and a needle tube 23 .

[0064] in:

[0065] The needle tube 23 is mounted on the Z-axis of the moving platform.

[0066] The curved needle tip 21 is arranged below the needle tube 23, and the curved needle tip 21 is locked on the needle tube 23 through the metal pad 22. The curvature of the curved needle tip 21 can be 75 degrees or 90 degrees. The diameter of the needle hole of the curved needle tip 21 is very small. Under the action of air pressure, the glue can be squeezed out from the curved needle tip 21 along the needle tube 23.

[0067] The metal pad 22 is screwed under the needle tube 23, and an opening is provided in the middle of the metal pad 22 so that the curved needle tip 21 can pass through it, and the curved needle tip 21 can be replaced quickly and conveniently. The material of the metal pad 22 is frosted, mainly to absorb the light source and facilitate manual tightening. The back 221 of the metal pad 22 is a silver reflective material back, which can effectively ensure that when the camera 121 collects the curved needle tip image of the curved needle tip 21, the background appears white under the illumination of the white coaxial light source provided by the coaxial light source 122, so as to reduce the noise as much as possible and facilitate binary analysis during image processing.

[0068] After the curved needle tip 21 passes through the opening in the middle of the metal pad 22, it can be buckled in the opening in the middle of the metal pad 22 and will not fall off. The metal pad 22 is screwed under the needle tube 23. By manually tightening it, the curved needle tip 21 can be fixedly installed under the needle tube 23 and connected to the needle tube 23, so that under the action of air pressure, glue can be squeezed out of the curved needle tip 21 along the needle tube 23.

[0069] In one embodiment, the processing device 30 is communicatively connected to the automated mechanism device 10 for sending control instructions to the automated mechanism device 10 and receiving a curved needle tip image transmitted by the automated mechanism device 10, performing image processing on the curved needle tip image based on a preset image processing algorithm, identifying the curved needle tip position in the curved needle tip image, and calibrating the curved needle tip based on the curved needle tip position in the curved needle tip image.

[0070] The specific processing process can refer to the relevant content of step S5 in the vision-based curved needle tip calibration method, which will not be repeated here.

[0071] For example, the processing device 30 is an electronic device with data processing and communication functions, and may be a personal computer, a server, or a workstation.

[0072] Based on the same concept, in one embodiment, Figure 2 As shown, the present invention provides a vision-based curved needle tip calibration method, which is applied to the vision-based curved needle tip calibration device 100 described in any of the above embodiments. The vision-based curved needle tip calibration device 100 includes an automated mechanism device, a rubber barrel curved needle tip device and a processing device; the vision-based curved needle tip calibration method includes:

[0073] S1. Mounting a plastic barrel bending needle tip device on an automated mechanism device, wherein the plastic barrel bending needle tip device includes a bending needle tip;

[0074] S2, the processing device sends a control instruction to the automation mechanism device;

[0075] S3, the automated mechanism moves according to the control instruction of the processing device, collects the curved needle tip image of the curved needle tip in the plastic barrel curved needle tip device, and transmits it to the processing device;

[0076] S5. The processing device performs image processing on the curved needle tip image based on a preset image processing algorithm, identifies the curved needle tip position in the curved needle tip image, and calibrates the curved needle tip based on the curved needle tip position in the curved needle tip image.

[0077] In this embodiment, a vision-based curved needle tip calibration method is provided, in which a glue barrel curved needle tip device is mounted on an automated mechanism device, the glue barrel curved needle tip device includes a curved needle tip; the automated mechanism device moves according to the control instructions of the processing device, and collects a curved needle tip image of the curved needle tip in the glue barrel curved needle tip device; the processing device performs image processing on the curved needle tip image based on a preset image processing algorithm, identifies the position of the curved needle tip in the curved needle tip image, and calibrates the curved needle tip based on the position of the curved needle tip in the curved needle tip image. Thus, by using the mechanical structure in conjunction with industrial vision technology to accurately identify and grasp the position of the curved needle tip for effective and accurate equipment calibration, it is further possible to provide a stable, reliable, more convenient and accurate correction operation for precise dispensing of product sides or gaps, improve dispensing accuracy, improve the process accuracy and accuracy of the dispensing equipment, and improve product quality; the automated mechanism moves according to the control instructions of the processing device, automatically collects the curved needle tip image of the curved needle tip in the glue barrel curved needle tip device, and the processing device automatically bends the needle tip image based on a preset image processing algorithm. The automated image processing and grasping process reduces manual intervention and improves the working efficiency of the dispensing equipment; the processing device calibrates the curved needle tip based on the curved needle tip position in the curved needle tip image, which can improve the reliability of the system. Thus, it can solve the existing problem of using the traditional straight curved needle tip calibration method to calibrate the curved needle tip, resulting in slow calibration speed, low efficiency, more manual intervention, and failure to provide more convenient and accurate correction operation for precise dispensing.

[0078] In one embodiment, in step S3, the automated mechanism device moves according to the control instruction of the processing device, collects the curved needle tip image of the curved needle tip in the plastic barrel curved needle tip device, and transmits it to the processing device, including:

[0079] S31, based on the angle of the mobile platform rotation axis, the plane where the rotation axis rotates one circle is evenly divided into a number of rotation positions at a preset rotation angle, wherein the preset rotation angle is the angle of the mobile platform rotation axis evenly divided by the number of rotation positions corresponding to the number of rotation positions. Specifically including:

[0080] S311, determine that the right direction of the rotation axis of the mobile platform is the rotation axis 0 position, and the clockwise direction is the positive direction.

[0081] S312. Under the premise that the calibration origin position XYZ axis remains unchanged, based on the angle of 360° rotation of the mobile platform rotation axis, the plane where the rotation axis rotates one circle is evenly divided into several rotation positions in the clockwise direction according to the preset rotation angle, wherein the preset rotation angle is the angle of the mobile platform rotation axis evenly divided by the number of corresponding rotation positions.

[0082] For example, the angle of one rotation of the mobile platform's rotation axis is 360°, which is evenly divided into 8 rotation positions in a clockwise direction on the plane where the rotation axis rotates one circle according to a preset rotation angle, wherein the preset rotation angle is the angle of one rotation of the mobile platform's rotation axis divided evenly by the number 8 corresponding to the 8 rotation positions, that is, the preset rotation angle is 45° (360° / 8=45°), which is evenly divided into 8 rotation positions in a clockwise direction on the plane where the rotation axis rotates one circle according to the preset rotation angle of 45°, namely 0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315°, a total of 8 rotation positions.

[0083] In the following embodiments of the present invention, the preset rotation angle is 45° and the plane where the rotation axis rotates one circle in the clockwise direction is evenly divided into 8 rotation positions. However, it can be understood that the preset rotation angle of the present invention is evenly divided into a plurality of rotation positions on the plane where the rotation axis rotates one circle in the clockwise direction, and the number of the plurality of rotation positions is not limited to 8, and the number can be an even number.

[0084] S32, adjusting camera parameters, using the camera to obtain the curved needle tip image at each rotation position, and transmitting it to the processing device; specifically including:

[0085] S321, rotating the rotating axis of the mobile platform to one of the rotating positions.

[0086] S322, adjust the mobile platform to the center of the camera, and debug the camera light source and camera focal length parameters to make the curved needle tip clear within the camera image field of view.

[0087] S323, respectively obtain the curved needle tip image within the camera image field of view, such as Figure 1 or Figure 2 shown.

[0088] Figure 3 This is a 75° curved tip image. Figure 4 This is a 90° bent tip image. The difference is Figure 3The curved needle tip is shorter. Figure 4 The curved needle tip is longer. Figure 3 The inner width of the white part of the curved needle tip is close to 0. Figure 4 The white part of the curved needle tip is the obvious inner width.

[0089] S324, repeat the above steps S321 to S323 until the curved needle tip image at each rotation position is obtained, and transmit the curved needle tip image at each rotation position to the processing device.

[0090] For example, the preset rotation angle is 45° and the plane where the rotation axis rotates one circle in the clockwise direction is evenly divided into 8 rotation positions. The mobile platform rotation axis is rotated to the first rotation position of the preset rotation angle, that is, the rotation position of the rotation angle of 0°, and the curved needle tip image at the first rotation position of the rotation angle of 0° is obtained; then the mobile platform rotation axis is rotated to the second rotation position of the preset rotation angle, that is, the rotation position of the rotation angle of 45°, and the curved needle tip image at the second rotation position of the rotation angle of 45° is obtained; then the mobile platform rotation axis is rotated to the third rotation position of the preset rotation angle, that is, the rotation position of the rotation angle of 90°, and the curved needle tip image at the third rotation position of the rotation angle of 90° is obtained; until the curved needle tip image at the ninth rotation position of the rotation angle of 315° is obtained.

[0091] In this embodiment, based on the angle of one rotation of the rotating axis of the mobile platform, the plane on which the rotating axis rotates one circle is evenly divided into several rotation positions at preset rotation angles. A camera is used to obtain the image of the curved needle tip at each rotation position and transmit it to a processing device. This can achieve automated image capture, reduce manual intervention, and improve the working efficiency of the dispensing equipment.

[0092] In one embodiment, Figure 2 As shown, the vision-based curved needle tip calibration method further includes: S4, the processing device determines the calibration relationship between the camera center and the mobile platform. Specifically including:

[0093] S41, move the mobile platform so that the rotation center of the rotation axis is aligned with the center position of the camera, the direction of the bending needle tip of the rotation axis is in the right direction, and record the current position as the origin position.

[0094] S42, adjust the camera image field of view to the best clarity, including:

[0095] In the camera continuous acquisition mode, adjust the light source brightness, camera exposure, gain, and contrast to the optimal values.

[0096] S43, using a camera to collect a curved needle tip image, and performing a 4*4 matrix camera calibration based on the curved needle tip image to determine a calibration relationship between the camera center and the mobile platform, including:

[0097] S431, creating a template image and matching parameters for the template image, specifically including:

[0098] Capture a single-frame curved needle tip image, draw a green rectangular frame on the curved needle tip image, and select the central area of ​​the curved needle tip in the curved needle tip image as the template image; draw a red rectangular frame on the curved needle tip image, and use the entire curved needle tip image as the search area image; match the template image with parameters, including search score, angle, scaling, noise, and contrast, to ensure successful and accurate matching every time, and mark the center position of the template with a red crosshair.

[0099] S432: Generate a number of rotation position physical coordinates at the current coordinate position of the mobile platform within the preset camera image field of view to form a trajectory list, specifically including:

[0100] Set the camera image field size to 10mm*10mm (close to but smaller than the real field size); generate 16 physical coordinates of 4 rows*4 columns up, down, left, and right at the current coordinate position of the mobile platform to form a track table. The maximum horizontal movement stroke does not exceed 10mm, and the maximum column movement stroke does not exceed 10mm. That is, the above 16 physical coordinates are all within the camera image field of view. The physical coordinates here only change the X and Y axes, but not the Z axis.

[0101] S433, based on the physical coordinates of the rotation positions in the trajectory list, obtaining the template center position coordinates corresponding to the physical coordinates of the rotation positions on the template map, and preliminarily generating the calibration relationship between the camera center and the mobile platform, specifically including:

[0102] Traverse the physical coordinates of the above 16 positions in the trajectory list, obtain a total of 16 positions from left to right and from top to bottom on the template map, and obtain the template center position coordinates of each position on the template map, and record the template center position coordinates corresponding to the physical coordinates of the above 16 positions mapped in the template map in turn.

[0103] S434, based on the template center position coordinates corresponding to the physical coordinates of the plurality of rotation positions on the template image, the camera calibration is calculated using the homography matrix method, specifically including:

[0104] Using the principle of least squares, through the mathematical calculation method of linear transformation and perspective matrix transformation, the above 16 physical coordinates are mapped to the corresponding template center coordinates in the template image to construct a set of equations to solve the homography matrix and perform camera calibration. This is the projection matrix from an image coordinate plane to another physical coordinate plane, which can map points on one plane to corresponding points on another plane. Specifically, the coordinates of 16 groups of corresponding points in the two images are known, and they are substituted into the corresponding equations to construct a set of equations to solve the homography matrix.

[0105] S435. Determine the real field size of the camera image field according to the pixel size of the entire curved needle tip image, repeat the above steps S432 and S434, and accurately generate the calibration relationship between the camera center and the mobile platform.

[0106] S436, verify the calibration relationship between the camera center and the mobile platform: select any feature point position on the curved needle tip image, and use the calibration relationship between the camera center and the mobile platform to determine the corresponding physical coordinate position.

[0107] In one embodiment, in step S5, the processing device performs image processing on the curved needle tip image based on a preset image processing algorithm, identifies the curved needle tip position in the curved needle tip image, and performs curved needle tip calibration based on the curved needle tip position in the curved needle tip image. Specifically, it includes:

[0108] S51, based on a preset image processing algorithm, searching for the curved needle tip from the curved needle tip image at each rotation position, and obtaining a curved needle tip search result. Specifically including:

[0109] S511, performing image processing on the curved needle tip image at a preset rotation angle position to obtain a processed curved needle tip image; comprising:

[0110] S5111. Use an image processing library (such as OpenCV) to extract the edge of the curved needle tip image at a preset rotation angle position, and perform grayscale processing to convert it into a grayscale space to obtain a binary image.

[0111] S5112. Use the Canny edge detection operator to perform image edge detection on the binary image to obtain an image edge group E.

[0112] Furthermore, in order to remove noise and refine edges, morphological operations (for example, morphological operations such as dilation, erosion, opening and closing operations) are used to obtain an optimized image edge group E.

[0113] S512. Based on the image edge group E, determine the position of the curved needle tip.

[0114] S5121. Determine the center point of the image. The center point of the image is a point with a pixel size of (width / 2, height / 2) of the entire curved needle tip image.

[0115] S5122. Draw a group of concentric circles with the center point of the image as the center. Each concentric circle intersects with the image edge group E to obtain a point set. Select point pairs with distances between the parameters [minimum distance, maximum distance] from the point set, and use the center of the point pair as the candidate point, so as to obtain candidate points for all concentric circles and form a candidate point set.

[0116] S5123, fit straight lines with candidate points in the candidate point set, discard straight lines that are shorter than the center point of the image, and retain straight lines that are longer than the center point of the image. In some embodiments, inner width and outer width are also provided to limit the straight lines that need to be retained.

[0117] S5124. Among the retained straight lines, the line segment whose two ends are farthest from the center point of the image is the approximate position of the curved needle tip.

[0118] S5125. Find the intersection of the straight line and the image edge group E at the approximate position of the curved needle tip, which is the position of the curved needle tip. At this time, draw a green cross line at the curved needle tip.

[0119] S513. Based on the curved needle tip image and the curved needle tip position, measure the parameters of the curved needle tip.

[0120] S5131, measuring the inner width and outer width of the curved needle tip, including: determining the inner and outer edges of the curved needle tip by edge analysis of the curved needle tip image, and then measuring the inner width and outer width. In the above S5123 step, the required edge line can be found more accurately.

[0121] S5132, measuring the length of the curved needle tip, including: using the straight line feature of the curved needle tip, extracting the distance from the end point of the curved needle tip as the starting point to the center point of the image from the curved needle tip image to obtain the length, and extending the curved needle tip with a green line based on the above step j.

[0122] S5133. Adjust the contrast and threshold of the curved needle tip image. Iron Blood: Dynamically adjust the contrast and threshold of the curved needle tip image according to different lighting conditions and the curved needle tip material to enhance the distinction between the curved needle tip and the background.

[0123] S514, based on the position of the curved needle tip, searching for the curved needle tip from the curved needle tip image to obtain a curved needle tip search result; specifically including:

[0124] S5141. Determine the selection area in the curved needle tip image, including: according to the size and shape of the curved needle tip, draw a green rectangular frame with an arrow, the direction of the arrow is the template angle direction, and the right direction is 0 degrees, determine the selection area, and avoid other interference factors from entering the detection range. The selection of the selection area is based on: at the origin position, in the camera continuous image acquisition mode, the rotation axis rotates 360 degrees, and the position of the curved needle tip does not exceed the green rectangular frame, wherein the selection area saves the pixel parameter data of the green rectangular frame, such as the center coordinates, length, width, and angle.

[0125] S5142, determining the direction of the curved needle tip in the curved needle tip image, including: determining the direction of the curved needle tip in the curved needle tip image by analyzing the shape characteristics and edge direction of the curved needle tip. The specific process can refer to the above steps S5121 to S5125, which will not be repeated here.

[0126] Center positioning: Calculate the center position of the image, and adjust the upper moving mechanism so that one end of the curved needle tip is always located in the center area of ​​the image. The specific process can be referred to the above step S5121, which will not be repeated here.

[0127] S5143. Based on the direction of the curved needle tip, searching for the curved needle tip in the curved needle tip image includes: using edge detection algorithm and shape analysis to identify the position of the curved needle tip in the curved needle tip image.

[0128] Specifically, after determining the direction of the curved needle tip, the curved needle tip image is binarized, and the edge detection algorithm and shape analysis of the curved needle tip image are used to extract the edge area, separate the curved needle tip from the curved needle tip image background, and the curved needle tip appears as a distinct black strip. The inner and outer edges of the black strip edge are captured to obtain the position of the curved needle tip in the curved needle tip image. In the process of separating the curved needle tip from the curved needle tip image background, it is necessary to clearly avoid obvious noise in the background image. If there are too many noise points, the position of the curved needle tip may be deviated, and the curved needle tip image must be clear.

[0129] S5144, according to the position of the curved needle tip in the curved needle tip image, marking the result of searching for the curved needle tip, including:

[0130] If the curved needle tip is successfully found in the curved needle tip image, the position of the curved needle tip is marked in the curved needle tip image using a marking symbol or color, and the corresponding pixel coordinate value is recorded so that the operator can intuitively understand the position of the curved needle tip. For example, a successful search is displayed with a green crosshair.

[0131] If the curved needle tip is not found in the curved needle tip image, a prompt message indicating that the curved needle tip search failed is marked so that timely measures can be taken to deal with it. For example, the curved needle tip search failure is displayed in red NG.

[0132] S52. Based on the result of finding the curved needle tip, calibrate the curved needle tip that is successfully found.

[0133] Specifically, after completing the search for the curved needle tip at several rotation positions (for example, 8 rotation positions) and recording the corresponding pixel coordinate values ​​of the successfully found curved needle tip, the successfully found curved needle tip can be calibrated.

[0134] In the present invention, 8 rotation positions are used as an example for explanation. After the bending needle tip rotation of 8 rotation positions is completed, the bending needle tip search results of 8 bending needle tips are obtained, and 8 bending needle tip points are obtained. These 8 bending needle tip points are used to fit an ellipse, and the position of the bending needle tip is inferred using the ellipse equation and the rotation angle to perform bending needle tip calibration.

[0135] The specific process is as follows:

[0136] Assuming that the center of the ellipse is the center of rotation, the position of the curved needle tip at any rotation angle can be calculated by the following formula.

[0137] 1) General form of ellipse equation: The general equation of an ellipse in a rectangular coordinate system is:

[0138] Ax 2 +Bxy+Cy 2 +Dx+Ey+F=0

[0139] 2) Rotation angle and coordinate transformation: Let the original point coordinates be (x, y), the rotation angle be θ, and the coordinates after rotation be (x', y'), then the coordinate transformation formula is:

[0140] x′=xcosθ-ysinθ

[0141] y′=xsinθ+ycosθ

[0142] 3) Coordinates of the bending needle tip and the ellipse equation: There are 8 bending needle tips. Assume that the original coordinates of the bending needle tip are (xi, yi), and the rotation angles are θ1, θ2…θ8, respectively. Then the coordinates of the rotating bending needle tip are (xi′, yi′), which satisfies the ellipse equation:

[0143]

[0144] Substituting the coordinate transformation formula into the ellipse equation, we get the equation for the coordinates of the original curved needle tip and the rotation angle:

[0145] A(x i cosθ i -y i sinθ j ) 2 +B(x i cosθ j-y i sinθ j )(x i sinθ j +y i cosθ j )

[0146] +C(x i sinθ j +y i cosθ j ) 2 +D(x i cosθ j -y i sinθ j )+E(x i sinθ j +y i cosθ j )+F=0

[0147] Further organized into:

[0148]

[0149] 4) Fitting the ellipse using the curved needle points: There are 8 curved needle points, that is, there are 8 equation groups:

[0150]

[0151] Convert it into matrix form:

[0152]

[0153] 5) Use the least squares method and other methods to solve the estimated values ​​of the coefficients A, B, C, D, E, and F in the elliptic equation.

[0154]

[0155] Once the coefficients of the ellipse equation are determined, the position of the curved needle tip can be inferred using the coordinate transformation formula based on the given rotation angle. The specific process is as follows:

[0156] ①, let the equation of the ellipse be:

[0157] (x 2 / a 2 )+(y 2 / b 2 )=1, where a and b are known coefficients.

[0158] ②. Let the rotation angle be θ.

[0159] ③. The coordinate transformation formula is as follows:

[0160] For the original coordinates (x, y), the rotated coordinates are (x', y'):

[0161] x'=x*cosθ-y*sinθ

[0162] y'=x*sinθ+y*cosθ

[0163] ④ Now it is known that the position of the bent needle tip after rotation is on the ellipse, which satisfies the following ellipse equation:

[0164] (x' 2 / a 2 )+(y' 2 / b 2 )=1

[0165] Substituting the coordinate transformation formula into the above ellipse equation, we get the following equation:

[0166]

[0167] Through the above equation, when the coefficients a and b of the ellipse equation and the rotation angle θ are known, the original coordinates (x, y) of the curved needle tip can be deduced. For example:

[0168] ⑤. Assuming coefficients A = 9, B = 0, C = 4, D = 0, E = 0, F = -36, the fitted ellipse equation is:

[0169] 9x 2 +4y 2 -36=0, after transformation we can get: x 2 / 4+y 2 / 9=1.

[0170] It can be seen that the center of the ellipse is (0,0), the major axis a=2, the minor axis b=3, and the ellipse rotates 45° counterclockwise around the origin. First, convert the angle 45° into radians θ=π / 4. After the rotation, the coordinates of the curved needle tip are (x',y')=(1,2). Then, according to the coordinate transformation formula, first calculate:

[0171]

[0172] Then calculate the value of x:

[0173]

[0174] Then calculate the value of y:

[0175]

[0176] Therefore, the original position of the curved needle tip is:

[0177] In summary, the above formula and method can be used to use the ellipse equation and the rotation angle to infer the position of the curved needle tip and calibrate the curved needle tip.

[0178] In this embodiment, based on the calibration relationship between the camera center and the mobile platform, the curved needle tip is searched from the curved needle tip image at each rotation position based on the preset image processing algorithm to obtain the result of searching for the curved needle tip; from the result of searching for the curved needle tip, the curved needle tip successfully found is calibrated using the ellipse equation and the rotation angle. Thus, the position and parameters of the curved needle tip can be precisely controlled to improve the dispensing accuracy, ensure the accuracy and stability of the dispensing, and improve the product quality; the automated image capture and processing process reduces manual intervention and improves the working efficiency of the dispensing equipment; in the calibration process, timely abnormal prompts can be used to avoid dispensing errors caused by inaccurate position of the curved needle tip or failure to find the curved needle tip, thereby improving the reliability of the system. Thus, by using the mechanical structure in conjunction with industrial vision technology to accurately identify and capture the position of the curved needle tip for effective and accurate equipment calibration, a stable, reliable, more convenient and accurate correction operation for precise dispensing of product sides or gaps can be further realized, thereby improving the process accuracy, accuracy and reliability of the dispensing equipment. This can solve the problem that the existing method of using the traditional straight curved needle tip calibration method to calibrate the curved needle tip causes slow calibration speed, low efficiency, much human intervention, and inability to provide more convenient and accurate correction operations for precision dispensing.

[0179] In one embodiment, the vision-based curved needle tip calibration method further includes: S6, after the curved needle tip calibration is completed, verifying the correctness of the curved needle tip calibration.

[0180] Specifically, after the calibration of the curved needle tip is completed, the curved tip calibration can be used to verify the correctness of the curved tip calibration. If the rotary axis does not have a zero return or limit signal, the zero return of the rotary axis depends on the direction of the template made by the curved needle tip, which makes the initial position of the curved needle tip more flexible in direction. According to the characteristics of the product placement, the user can configure different rotation values ​​of the rotary axis. The dispensing path generated by the trajectory can clearly show the quality of the dispensing, thereby verifying the correctness of the curved needle tip calibration.

[0181] It should be noted that the above-mentioned vision-based curved needle tip calibration method embodiment and the above-mentioned vision-based curved needle tip calibration device embodiment belong to the same concept, and the specific implementation process is detailed in the vision-based curved needle tip calibration device embodiment, and the technical features in the vision-based curved needle tip calibration device embodiment are correspondingly applicable in the above-mentioned vision-based curved needle tip calibration method embodiment, which will not be repeated here.

[0182] It should be noted that, in this article, the terms "include", "comprises" or any other variations 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. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0183] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may also be combined, the steps may be implemented in any order, and there are many other changes in different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity. Although the present invention has been described in detail with reference to the above embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vision-based curved needle tip calibration device, characterized in that: It includes an automated mechanism, a glue barrel bending needle tip device and a processing device, wherein: The glue barrel curved needle tip device is mounted on the automated mechanism device and includes a curved needle tip; The processing device is communicatively connected with the automation mechanism device, and is used to send control instructions to the automation mechanism device; The automated mechanism device is used to move according to the control instruction, collect the curved needle tip image of the curved needle tip, and transmit it to the processing device; The processing device is also used to perform image processing on the curved needle tip image based on a preset image processing algorithm, identify the curved needle tip position in the curved needle tip image, and calibrate the curved needle tip based on the curved needle tip position in the curved needle tip image.

2. The vision-based curved needle tip calibration device according to claim 1, characterized in that: The automated mechanism device comprises a mobile platform and a camera device, wherein: The mobile platform is located on the upper part of the camera device and is used to move the rotation center of the rotation axis to align with the camera center position of the camera device according to the control instruction; The photographic device is used to collect the curved needle tip image of the curved needle tip and transmit it to the processing device.

3. The vision-based curved needle tip calibration device according to claim 2, characterized in that: The mobile platform includes an X-axis, a Y-axis, a Z-axis and an R-axis, wherein the effective stroke of the Z-axis is 0 to 360 degrees; the R-axis is a rotation axis with an effective stroke of 0 to 360 degrees, the clockwise direction of the rotation axis is set as the positive direction, and the curved needle tip is rotated to the right direction to be set as the 0 position of the rotation axis.

4. The vision-based curved needle tip calibration device according to claim 3, characterized in that: The photographic device comprises a camera and a coaxial light source, wherein: The coaxial light source provides a white coaxial light source for the camera; The camera is used to collect the curved needle tip image of the curved needle tip under the illumination of the white coaxial light source provided by the coaxial light source, and transmit it to the processing device.

5. The vision-based curved needle tip calibration device according to claim 4, characterized in that: The plastic barrel curved needle tip device also includes a metal pad and a needle tube, wherein: The needle tube is mounted on the Z axis of the mobile platform; The curved needle tip is arranged below the needle tube, and the curved needle tip is locked on the needle tube through the metal pad; The metal pad is screwed under the needle tube, and an opening is arranged in the middle of the metal pad so that the curved needle tip can pass through the opening and be buckled in the middle of the metal pad.

6. The vision-based curved needle tip calibration device according to claim 5, characterized in that: The back of the metal pad is made of a silver reflective material, which ensures that when the camera collects the curved needle tip image of the curved needle tip, the background appears white under the illumination of the white coaxial light source provided by the coaxial light source, so as to reduce noise.

7. A method for calibration of curved needle tip based on vision, characterized in that: The vision-based curved needle tip calibration device applied to any one of claims 1 to 6, the vision-based curved needle tip calibration device comprising an automated mechanism device, a rubber barrel curved needle tip device and a processing device; The vision-based curved needle tip calibration method comprises: The glue barrel bending needle tip device is mounted on the automated mechanism device, wherein the glue barrel bending needle tip device includes a bending needle tip; The processing device sends a control instruction to the automation mechanism device; The automated mechanism device moves according to the control instruction, collects the curved needle tip image of the curved needle tip, and transmits it to the processing device; The processing device performs image processing on the curved needle tip image based on a preset image processing algorithm, identifies the curved needle tip position in the curved needle tip image, and calibrates the curved needle tip based on the curved needle tip position in the curved needle tip image.

8. The method for calibration of curved needle tip based on vision according to claim 7, characterized in that: The automated mechanism moves according to the control instruction, collects the curved needle tip image of the curved needle tip, and transmits it to the processing device, including: Based on the angle of one rotation of the rotating axis of the mobile platform, the plane on which the rotating axis rotates one circle is evenly divided into a number of rotation positions at a preset rotation angle, wherein the preset rotation angle is the angle of one rotation of the rotating axis of the mobile platform evenly divided by the number of corresponding rotation positions; The camera parameters are adjusted, and the image of the curved needle tip at each rotation position is acquired by using the camera and transmitted to the processing device.

9. The method for calibration of curved needle tip based on vision according to claim 7, characterized in that: The vision-based curved needle tip calibration method also includes the processing device determining a calibration relationship between a camera center and a mobile platform.

10. The method for calibration of curved needle tip based on vision according to claim 9, characterized in that: The processing device performs image processing on the curved needle tip image based on a preset image processing algorithm, identifies the curved needle tip position in the curved needle tip image, and performs curved needle tip calibration based on the curved needle tip position in the curved needle tip image, including: Based on a preset image processing algorithm and a calibration relationship between the camera center and the mobile platform, the curved needle tip is searched from the curved needle tip image at each rotation position to obtain a curved needle tip search result; Based on the result of searching for the curved needle tip, the curved needle tip successfully found is calibrated.

Images