A decal robot vision processing system and method

The decal robot vision processing system, which combines a multi-CCD vision camera and a six-axis robot, along with a cylinder and nozzle design, solves the problems of low efficiency and low precision in existing systems, and achieves efficient and accurate decal picking, placing and applying.

CN117262419BActive Publication Date: 2026-03-17ANHUI UNIVERSITY OF TECHNOLOGY
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-11-08
Publication Date
2026-03-17

AI Technical Summary

Technical Problem

Existing visual processing systems for decal robots are inefficient and lack precision in mass industrial production, and are prone to causing wrinkles and damage to the decals.

Method used

By combining a multi-CCD vision camera with a six-axis robot, image analysis is performed using a minimum variance iterative circle fitting algorithm. Combined with cylinder and nozzle design, precise paper picking and applying is achieved, and a filter steel mesh is used to reduce instantaneous impact.

Benefits of technology

It improves the precision and efficiency of decal production, avoids damage and wrinkles, and enables efficient assembly line production.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117262419B_ABST
    Figure CN117262419B_ABST
Patent Text Reader

Abstract

This invention belongs to the field of robotics, specifically relating to a visual processing system and method for a decal robot. The visual processing system for the decal robot is scientifically and rationally designed. The decal is moved to the desired photographing position by a first flipping mechanism. Images of the decal are captured and scanned by a first CCD vision camera and a third CCD vision camera. The center of the decal is determined using a minimum variance iterative circle fitting algorithm, and the decal position information is then sent to a first six-axis robot and a second six-axis robot for decal removal. A second CCD vision camera and a fourth CCD vision camera photograph and analyze the position of the bottle to be decaled, and the position information is sent to the first and second six-axis robots for decal application. The robot's end effector has a special air nozzle, preventing the flexible, wet decal from tearing, wrinkling, or shifting after application. The second flipping mechanism then replaces the bottle, thus achieving efficient assembly line operation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of robotics technology, specifically relating to a visual processing system and method for a decal robot. Background Technology

[0002] With economic development and improved living standards, people are paying increasing attention to product quality and aesthetic appeal, and labels are one of the best representations of product information and appearance. Label application primarily relies on automated labeling equipment. During the labeling process, the accuracy of positioning the label and the precision of the labeling equipment directly determine its efficiency. Machine vision systems, on the other hand, enhance production flexibility and automation. They are commonly used to replace manual labor in hazardous work environments or where human vision is insufficient. Furthermore, machine vision can locate and identify products on the fastest production lines, ensuring both quality and quantity in production tasks.

[0003] Existing vision processing systems are being used in decal robots, but their application is still imperfect. These systems have several shortcomings: firstly, they are inefficient and lack precision in large-scale industrial production; secondly, they cause wrinkles and damage when handling flexible decals. Therefore, structural optimization and improvement are necessary. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned problems existing in the conventional technology and to provide a visual processing system and method for decal robots.

[0005] To achieve the above-mentioned technical objectives and effects, the present invention is implemented through the following technical solution:

[0006] This invention provides a visual processing system for decal robots, including a worktable on which a first flipping mechanism and four CCD vision cameras are mounted. The first flipping mechanism has a decal hopper, and the four CCD vision cameras are a first CCD vision camera, a second CCD vision camera, a third CCD vision camera, and a fourth CCD vision camera. A second flipping mechanism for loading bottles to be decaled is mounted at the edge of the worktable, and a first six-axis robot and a second six-axis robot are mounted on the second flipping mechanism. The first and second six-axis robots can retrieve decals from the decal hopper and then apply them to the bottles loaded on the second flipping mechanism. The first and second six-axis robots work alternately.

[0007] Furthermore, in the above-mentioned decal robot vision processing system, a decal bin is fixed above the first flipping mechanism. The decal bin is secured with a clamp and the decal is replaced by the first flipping mechanism.

[0008] Furthermore, in the aforementioned decal robot vision processing system, the second flipping mechanism drives a sprocket via a servo motor to replace the wine bottle to be decaled placed on top of the second flipping mechanism.

[0009] Furthermore, in the aforementioned decal robot vision processing system, the first six-axis robot is equipped with a first cylinder and a second cylinder that can move up and down on its first rotating axis. The movable end of the first cylinder is equipped with a first nozzle, and the movable end of the second cylinder is equipped with a second nozzle. The first nozzle and the second nozzle can perform the operation of picking up and applying decals through the action of the air valve component.

[0010] Furthermore, in the above-mentioned visual processing system for the decal robot, a third cylinder and a fourth cylinder capable of moving up and down are installed on the second rotating axis of the second six-axis robot. A third nozzle is installed on the movable end of the third cylinder, and a fourth nozzle is installed on the movable end of the fourth cylinder. The third and fourth nozzles can perform the operation of picking up and applying decals through the action of the air valve component.

[0011] Furthermore, in the aforementioned decal robot vision processing system, the ends of the four air nozzles are all equipped with filter steel meshes with good rigidity and positioning function. A filter nylon mesh is installed above the filter steel meshes, which can reduce the instantaneous impact when picking up the decals, prevent the decals from being torn or wrinkled, and also prevent the decals from being applied off-center.

[0012] Furthermore, in the aforementioned decal robot vision processing system, after the CCD vision camera takes a picture, it uploads it to the host computer, and the image is analyzed using the decal circle center algorithm. The decal circle center algorithm uses the minimum variance iterative circle fitting algorithm: in polar coordinates, the center coordinates of the fitted circle are set to (m, α), and the coordinates of the points on the circle are (θ). i , ρ i ), where i is the index of the discrete point obtained; the equation of the circle in polar coordinates is ρ 2 i -2m i ρ i cos(α-θ i )+m 2 =R 2 To obtain the initial m and α, refer to the least squares circle fitting method:

[0013]

[0014]

[0015] The obtained (m, α) is used as the coordinates of the fitted center of the decal.

[0016] The present invention also provides a visual processing method for a decal robot, which is based on the above-mentioned visual processing system for a decal robot. The method includes the following steps:

[0017] S1. The first flipping mechanism transports the decal to the position to be photographed. After stabilization, the sensor triggers the first CCD vision camera and the second CCD vision camera to take pictures and collect the decal position information in the decal bin (1).

[0018] S2. The host computer processes the images captured by the camera and determines whether there is decal paper in the decal paper bin. If there is no decal paper, return to step S1. If there is decal paper, the first six-axis robot and the second six-axis robot retrieve the decal paper according to the position information.

[0019] S3. The second flipping mechanism transports the bottle to be pasted to the position to be photographed. The sensor triggers the third CCD vision camera and the fourth CCD vision camera to collect the position information of the bottle to be pasted (9).

[0020] S4. The host computer processes the images captured by the vision camera and determines whether the wine bottle to be decorated has already been decorated. If it has been decorated, step S3 will be repeated; otherwise, the first six-axis robot and the second six-axis robot will decorate the wine bottle according to the position information of the wine bottle.

[0021] S5. The decal application is completed, and the second flipping mechanism transports a new bottle to be decaled.

[0022] The beneficial effects of this invention are:

[0023] 1. The visual processing system of the decal robot of this invention is reasonably designed. The wine bottle to be decaled and the decal are transported to a fixed position in sequence by the flipping mechanism. Then, the CCD vision camera takes a picture and uploads it to the host computer. The minimum variance iterative circle fitting algorithm is used for image analysis to quickly and accurately obtain relevant position information, so as to accurately and stably pick up and apply the decal. The robot has two air nozzles at the end, which can perform two decal operations at the same time, thereby achieving efficient assembly line production.

[0024] 2. The visual processing system of this invention for decal robots has a vision camera fixedly installed on the top of the equipment's worktable to collect position information of the decals and the substitute wine bottle, thereby making the decal effect more accurate. It identifies the image of the surface to determine whether there is material present or absent, and then the flipping mechanism replenishes the material in a timely manner.

[0025] 3. The air nozzle part of the visual processing system of the decal robot of the present invention has the function of sucking up flexible wet decal paper. The end of the air nozzle can reduce the instantaneous impact when picking up the drawing paper, avoid damage and wrinkles to the decal paper, and also prevent the decal paper from being applied off-center.

[0026] Of course, any product implementing this invention does not necessarily need to achieve all of the above advantages at the same time. Attached Figure Description

[0027] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a schematic diagram of the overall structure of the visual processing system for the decal robot of the present invention;

[0029] Figure 2 This is a schematic diagram of the structure of the six-axis robot in this invention;

[0030] Figure 3 This is a schematic diagram of the decal holder in this invention;

[0031] Figure 4 This is a flowchart of the visual processing method for the decal robot of the present invention;

[0032] In the attached diagram, the component numbers are as follows:

[0033] 1-Paper hopper, 2-First flipping mechanism, 3-Third CCD vision camera, 4-Fourth CCD vision camera, 5-First CCD vision camera, 6-Equipment workbench, 7-Second CCD vision camera, 8-Second flipping mechanism, 9-Bottle to be coated, 10-Second six-axis robot, 101-First rotary axis, 102-First cylinder, 103-Second cylinder, 104-First nozzle, 105-Second nozzle, 11-First six-axis robot, 111-Second rotary axis, 112-Third cylinder, 113-Fourth cylinder, 114-Third nozzle, 115-Fourth nozzle. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] The specific embodiments of the present invention are as follows:

[0036] like Figures 1 to 4As shown, this embodiment provides a visual processing system for a decal robot, including a worktable 6. A first flipping mechanism 2 and four CCD vision cameras are mounted on the worktable 6. A decal material bin 1 is mounted on the first flipping mechanism 2. The four CCD vision cameras are a first CCD vision camera 5, a second CCD vision camera 7, a third CCD vision camera 3, and a fourth CCD vision camera 4. A second flipping mechanism 8 for loading bottles 9 to be decaled is mounted at the edge of the worktable. A first six-axis robot 11 and a second six-axis robot 10 are mounted on the second flipping mechanism 8. The first six-axis robot 11 and the second six-axis robot 10 can pick up decals from the decal material bin 1 and then apply them to the bottles 9 loaded on the second flipping mechanism 8. The first six-axis robot 11 and the second six-axis robot 10 work alternately.

[0037] In this embodiment, the decal bin 1 is fixed above the first flipping mechanism 2. The decal bin 1 is fixed with a clamp to fix the decal and the decal is replaced by the first flipping mechanism 2.

[0038] In this embodiment, the second flipping mechanism 8 uses a servo motor to drive a sprocket, thereby replacing the wine bottle 9 to be labeled placed on top of the second flipping mechanism 8.

[0039] In this embodiment, a first cylinder 102 and a second cylinder 103 capable of moving up and down are installed on the first rotating axis 101 of the first six-axis robot 10. A first air nozzle 104 is installed on the movable end of the first cylinder 102, and a second air nozzle 105 is installed on the movable end of the second cylinder 103. The first air nozzle 104 and the second air nozzle 105 can perform the operation of picking up and applying decals through the action of the air valve component.

[0040] In this embodiment, a third cylinder 112 and a fourth cylinder 113 capable of moving up and down are installed on the second rotating axis 111 of the second six-axis robot 11. A third air nozzle 114 is installed on the movable end of the third cylinder 112, and a fourth air nozzle 115 is installed on the movable end of the fourth cylinder 113. The third air nozzle 114 and the fourth air nozzle 115 can perform the operation of picking up and applying decals through the action of the air valve component.

[0041] In this embodiment, the ends of the four air nozzles are all equipped with filter steel mesh 106 with good rigidity and positioning function. A filter nylon mesh 107 is installed above the filter steel mesh 106, which can reduce the instantaneous impact when picking up the drawing paper, avoid damage and wrinkles to the drawing paper, and also prevent the drawing paper from being misaligned.

[0042] In this embodiment, after the CCD vision camera takes a picture, it uploads it to the host computer, and the image is processed using the "circle center algorithm" for image analysis. The circle center algorithm uses the minimum variance iterative circle fitting algorithm: in polar coordinates, the center coordinates of the fitted circle are set to (m, α), and the coordinates of the points on the circle are (θ). i , ρi ), where i is the index of the discrete point obtained. The equation of the circle in polar coordinates is ρ. 2 i -2m i ρ i cos(α-θ i )+m 2 =R 2 To obtain the initial m and α, refer to the least squares circle fitting method:

[0043]

[0044]

[0045] The obtained (m, α) is used as the coordinates of the fitted center of the decal.

[0046] This embodiment also provides a visual processing method for a decal robot, implemented based on the aforementioned visual processing system for a decal robot. The method includes the following steps:

[0047] S1. The first flipping mechanism 2 transports the decal to the position to be photographed. After stabilization, the sensor triggers the first CCD vision camera 5 and the second CCD vision camera 7 to take pictures and collect the decal position information in the decal bin 1.

[0048] S2. The host computer processes the images captured by the camera and determines whether there is decal paper in the decal paper bin 1. If there is no decal paper, return to step S1. If there is decal paper, the first six-axis robot 10 and the second six-axis robot 11 retrieve the decal paper according to the position information.

[0049] S3. The second flipping mechanism 8 transports the bottle 9 to be pasted to the position to be photographed. The sensor triggers the third CCD vision camera 3 and the fourth CCD vision camera 4 to collect the position information of the bottle 9 to be pasted.

[0050] S4. The host computer processes the images captured by the vision camera and determines whether the decals on the bottle 9 to be decaled have been applied. If the decals have been applied, step S3 will be repeated; otherwise, the first six-axis robot 10 and the second six-axis robot 11 will apply the decals according to the bottle position information.

[0051] S5. The decal application is completed, and the second flipping mechanism 8 transports a new bottle to be decaled 9.

[0052] In this embodiment, the decal is moved to the position to be photographed by the first flipping mechanism 2. The first CCD vision camera 5 and the third CCD vision camera 3 capture and scan the image of the decal, and send the decal position information to the first six-axis robot 10 and the second six-axis robot 11 respectively for decal removal. The second CCD vision camera 7 and the fourth CCD vision camera 4 photograph and analyze the position of the bottle 9 to be decaled, and send the position information to the first six-axis robot 10 and the second six-axis robot 11 respectively for decal application. After the decal is applied, the second flipping mechanism 8 replaces the bottle 9 to be decaled, thus achieving efficient assembly line operation.

[0053] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to specific implementations. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A decal robot vision processing system comprising a device table (6), characterized in that: The device workbench (6) is provided with a first turnover mechanism (2) and four CCD visual cameras; the first turnover mechanism (2) is provided with a paper flower bin (1), and the four CCD visual cameras are a first CCD visual camera (5), a second CCD visual camera (7), a third CCD visual camera (3) and a fourth CCD visual camera (4) respectively; a second turnover mechanism (8) for loading a bottle to be pasted (9) is arranged at the edge of the device workbench (6), and the second turnover mechanism (8) is provided with a first six-axis robot (11) and a second six-axis robot (10); the first six-axis robot (11) and the second six-axis robot (10) can take paper flowers in the paper flower bin (1) and then paste the paper flowers on the bottle to be pasted (9) loaded on the second turnover mechanism (8), and the first six-axis robot (11) and the second six-axis robot (10) work alternately. The ends of the four air nozzles are provided with filter steel meshes (106) with good rigidity and positioning function, and the filter steel meshes (106) are provided with filter nylon meshes (107) above, which can slow down the instantaneous impact of taking paper flowers, avoid damage and wrinkles of the paper flowers, and also avoid offsetting of the paper flowers. After the CCD visual camera takes a picture, it is uploaded to the host computer, and image analysis is performed using a paper circle center algorithm. The paper circle center algorithm uses a minimum variance iterative circle fitting algorithm: in polar coordinates, the center coordinates of the fitted circle are set as (m, a), and the coordinates of the points on the circle are (r i, a i ) , ), where i is the serial number of the obtained discrete points; the equation of the circle in the polar coordinate system is ; to obtain the initial m and a, the least squares circle fitting method is referred to: ; ; The obtained (m, α) is taken as the fitting center coordinate of the paper flower.

2. The decal robot vision processing system of claim 1, wherein: The first turnover mechanism (2) is fixed above the paper flower bin (1), the paper flower bin (1) fixes the paper flower through a clamp, and the paper flower is replaced through the first turnover mechanism (2).

3. The decal robot vision processing system of claim 2, wherein: The second turnover mechanism (8) is driven by a servo motor to replace the bottle to be pasted (9) placed on the top of the second turnover mechanism (8).

4. The decal robot vision processing system of claim 3, wherein: The first rotating shaft (101) of the first six-axis robot (11) is provided with a first air cylinder (102) and a second air cylinder (103) capable of moving up and down, the movable end of the first air cylinder (102) is provided with a first air nozzle (104), the movable end of the second air cylinder (103) is provided with a second air nozzle (105), and the first air nozzle (104) and the second air nozzle (105) can realize the operations of sucking and pasting the paper flower through the action of the air valve component.

5. The decal robot vision processing system of claim 4, wherein: The second rotating shaft (111) of the second six-axis robot (10) is provided with a third air cylinder (112) and a fourth air cylinder (113) capable of moving up and down, the movable end of the third air cylinder (112) is provided with a third air nozzle (114), the movable end of the fourth air cylinder (113) is provided with a fourth air nozzle (115), and the third air nozzle (114) and the fourth air nozzle (115) can realize the operations of sucking and pasting the paper flower through the action of the air valve component.

6. A decal robot vision processing method, implemented based on the decal robot vision processing system of any one of claims 1-5, characterized in that, The method comprises the following steps: S1, the first turnover mechanism (2) transports the paper flower to a position to be photographed, waits for stabilization, and then triggers the first CCD visual camera (5) and the second CCD visual camera (7) to take a photograph, and collects the position information of the paper flower in the paper flower bin (1); The ends of the four air nozzles are provided with filter steel meshes (106) with good rigidity and positioning function, and the filter steel meshes (106) are provided with filter nylon meshes (107) above, which can slow down the instantaneous impact of taking paper flowers, avoid damage and wrinkles of the paper flowers, and also avoid offsetting of the paper flowers. S2, the host computer processes the pictures collected by the camera, judges whether there is paper in the paper warehouse (1), if not, return to step S1; if there is paper, the first six-axis robot (11) and the second six-axis robot (10) take the paper according to the position information; S3, the second turnover mechanism (8) transports the wine bottle (9) to be pasted to the position to be photographed, and the sensor triggers the third CCD vision camera (3) and the fourth CCD vision camera (4) to collect the position information of the wine bottle (9) to be pasted; S4, the host computer processes the pictures collected by the vision camera, judges whether the wine bottle (9) to be pasted has been pasted with paper, if yes, repeat step S3; otherwise, the first six-axis robot (11) and the second six-axis robot (10) paste the paper according to the position information of the wine bottle; S5, the paper pasting work is finished, and the second turnover mechanism (8) transports the new wine bottle (9) to be pasted.

Citation Information

Patent Citations

  • Datum point positioning method based on machine vision

    CN103235939A

  • Product labeling system

    CN116986115A

  • Automatic marking and surface-mounting mechanism of electronic component

    CN212891377U