An automatic visual detection device based on an AI algorithm

Abstract

The utility model provides a kind of automatic visual inspection device based on AI algorithm.The automatic visual inspection device based on AI algorithm includes: bottom plate;Detection table, the detection table is fixedly connected at the top of bottom plate, the top of the detection table is equipped with arc detection bin, the inner wall of the arc detection bin is equipped with conveying assembly at the bottom, the inside of the arc detection bin is equipped with adjusting assembly at the position close to both sides.The automatic visual inspection device based on AI algorithm provided by the utility model, by the design, large field of view low magnification lens is used to ensure detection efficiency, small field of view high magnification lens is used to ensure detection accuracy, the collaborative working mode of "quick screening, accurate review" is realized, unnecessary high magnification shooting area is reduced, data processing amount is reduced, the operation efficiency of AI algorithm is improved, applicable to the scene that both requires rapid batch detection and needs to identify tiny defects, such as mobile phone screen, precision mold etc.

Description

Technical Field

[0001] This utility model relates to the field of visual inspection technology, and in particular to an automated visual inspection device based on AI algorithms. Background Technology

[0002] Visual inspection is the use of machines to replace human eyes for measurement and judgment. Visual inspection involves using machine vision products to convert the captured target into image signals, which are then transmitted to a dedicated image processing system. Based on pixel distribution and information such as brightness and color, these signals are converted into digital signals. The image system performs various calculations on these signals to extract the target's features, and then controls the on-site equipment based on the judgment results.

[0003] AI-based automated visual inspection devices are automated inspection equipment that integrates machine vision technology and artificial intelligence algorithms. They can acquire image information of the object being inspected through image acquisition devices, and then use AI algorithms to analyze, process and identify the image, thereby automatically determining whether the object being inspected has defects, whether the size is qualified, and whether the appearance meets the standards, thus achieving non-contact, high-efficiency and high-precision automatic inspection of products or workpieces.

[0004] A single camera has a limited field of view. If a small field of view high magnification lens is used, the detection accuracy is high but the range is small, and the speed is slow when detecting large workpieces. If a large field of view low magnification lens is used, the detection speed is fast but the accuracy is insufficient, making it difficult to identify tiny defects. It is impossible to balance efficiency and accuracy.

[0005] Therefore, it is necessary to provide an automated visual inspection device based on AI algorithms to solve the above-mentioned technical problems. Utility Model Content

[0006] This invention provides an automated visual inspection device based on AI algorithms, which solves the problems of limited field of view of a single camera making it difficult to identify minute defects and the inability to balance efficiency and accuracy.

[0007] To solve the above-mentioned technical problems, the automated visual inspection device based on AI algorithm provided by this utility model includes: a base plate;

[0008] A testing platform is fixedly connected to the top of a base plate. An arc-shaped testing chamber is installed on the top of the testing platform. A conveying assembly is installed at the bottom of the inner wall of the arc-shaped testing chamber. Adjustment assemblies are installed near both sides inside the arc-shaped testing chamber. Each adjustment assembly includes a limiting structure, an external gear structure, a docking structure, and a fixing plate. The docking structure is used to rotatably connect the external gear structure to the inside of the limiting structure. Adjustment structures are installed on the inner surfaces of the two adjustment assemblies. A second drive structure is installed at one end of each adjustment structure. A first mounting plate is installed at the bottom of one adjustment structure, and a large field-of-view low-magnification lens is installed at the bottom of the first mounting plate. A small field-of-view high-magnification lens is installed at the bottom of the other adjustment structure via a second mounting plate.

[0009] Two drive components are respectively installed on the top of the arc-shaped detection chamber near both ends, and drive gears are installed at the output ends of both drive components;

[0010] The inner surfaces of the fixed plate and the docking structure are connected, the inner surfaces of the limiting structure and the arc-shaped detection chamber are connected, the external gear structure and the drive gear are meshed, and the two ends of the docking structure and the two ends of the inner wall of the limiting structure are slidably connected by ball bearings, which can increase the stability of the rotation of the external gear structure. The adjustment structure includes a bracket and an adjustment head. The second drive structure provides rotational driving force for the adjustment head, which drives the mounting plate to adjust the angle and improve the detection range. The bottom of the mounting plate is equipped with an infrared sensor for auxiliary positioning.

[0011] Preferably, a feeding assembly is installed on both sides of the testing table. The feeding assembly includes a support frame and a placement plate. The support frame is used to install the placement plate on both sides of the testing table.

[0012] The placement plate is used to place the workpiece.

[0013] Preferably, a control box is installed on the top of the base plate, a door is rotatably connected to the front of the control box, and an operation screen is installed on the front of the testing platform;

[0014] The control box is equipped with a lock on the door. Inside the control box are installed power switches and controllers to control the operation of the equipment. The operation panel allows manual control of the equipment and setting of operating parameters.

[0015] Preferably, a protective component is installed on the top of the arc-shaped detection chamber. The protective component includes a magnetic frame and a protective frame, and the magnetic frame is used to install the protective frame on the top of the arc-shaped detection chamber.

[0016] Preferably, the drive assembly includes a protective housing, a drive component, and an angle sensor, wherein the protective housing is used to mount the drive component that provides rotational driving force;

[0017] The drive unit has a self-locking function, and the angle sensor, together with the drive unit controller, can precisely control the rotation speed and rotation angle.

[0018] Preferably, the conveying assembly includes a first drive structure, a rotating shaft, and a conveyor belt, the rotating shaft being used to mount the conveyor belt inside the arc-shaped detection chamber.

[0019] Compared with related technologies, the automated visual inspection device based on AI algorithms provided by this utility model has the following beneficial effects:

[0020] This invention provides an automated visual inspection device based on AI algorithms. To improve the accuracy of the device, two adjustment components are installed inside the arc-shaped inspection chamber. A small field-of-view high-magnification lens is mounted on the adjustment structure of one component via a second mounting plate, while a large field-of-view low-magnification lens is mounted on the adjustment structure of the other component via a first mounting plate. A drive component rotates the external gear structure, adjusting the angles of the large field-of-view low-magnification lens and the small field-of-view high-magnification lens. The second drive structure and adjustment structure further adjust the left and right angles of the lenses, increasing the inspection area. This design utilizes a large field-of-view low-magnification lens to ensure inspection efficiency and a small field-of-view high-magnification lens to ensure inspection accuracy, achieving a collaborative working mode of "rapid screening and precise verification." It reduces unnecessary high-magnification shooting areas, lowers data processing volume, and improves the computational efficiency of the AI ​​algorithm. It is suitable for scenarios requiring both rapid batch inspection and identification of minute defects, such as mobile phone screens and precision molds. Attached Figure Description

[0021] Figure 1 A schematic diagram of a preferred embodiment of the AI-based automated visual inspection device provided by this utility model;

[0022] Figure 2 A schematic diagram of the adjustment component is provided for this utility model;

[0023] Figure 3 A schematic diagram of the structure of a large field-of-view, low-magnification lens for this utility model;

[0024] Figure 4 Provided for this utility model Figure 3 An enlarged view of point A shown;

[0025] Figure 5 A schematic diagram of the drive component is provided for this utility model.

[0026] The diagram is labeled as follows: 1. Base plate, 2. Control box, 3. Box door, 4. Detection table, 5. Feeding assembly, 501. Support frame, 502. Placement plate, 6. Conveying assembly, 601. First drive structure, 602. Rotating shaft, 603. Conveyor belt, 7. Arc-shaped detection chamber, 8. Protective assembly, 801. Magnetic frame, 802. Protective frame, 9. Operation panel, 10. Adjustment assembly, 101. Limiting structure, 102. External gear structure, 103. Docking structure, 104. Fixing plate, 11. Large field of view low magnification lens, 12. First mounting plate, 13. Drive gear, 14. Drive assembly, 141. Protective shell, 142. Drive component, 143. Angle sensor, 15. Through-hole, 16. Second drive structure, 17. Small field of view high magnification lens, 18. Second mounting plate, 19. Adjustment structure. Detailed Implementation

[0027] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0028] Please refer to the following: Figure 1 , Figure 2 , Figure 3 , Figure 4 and Figure 5 ,in, Figure 1 A schematic diagram of a preferred embodiment of the AI-based automated visual inspection device provided by this utility model; Figure 2 A schematic diagram of the adjustment component is provided for this utility model; Figure 3 A schematic diagram of the structure of a large field-of-view, low-magnification lens for this utility model; Figure 4 Provided for this utility model Figure 3 An enlarged view of point A shown;

[0029] Figure 5 A structural schematic diagram of the driving component is provided for this utility model. The automated visual inspection device based on AI algorithms includes: a base plate 1;

[0030] A testing platform 4 is fixedly connected to the top of the base plate 1. An arc-shaped testing chamber 7 is installed on the top of the testing platform 4. A conveying assembly 6 is installed on the bottom of the inner wall of the arc-shaped testing chamber 7. Adjustment assemblies 10 are installed near both sides inside the arc-shaped testing chamber 7. Each adjustment assembly 10 includes a limiting structure 101, an external gear structure 102, a docking structure 103, and a fixing plate 104. The docking structure 103 is used to rotatably connect the external gear structure 102 to the inside of the limiting structure 101. Adjustment structures 19 are installed on the inner surfaces of the two adjustment assemblies 10. A second drive structure 16 is installed at one end of each of the two adjustment structures 19. A first mounting plate 12 is installed at the bottom of one of the adjustment structures 19. A large field of view low magnification lens 11 is installed at the bottom of the first mounting plate 12. A small field of view high magnification lens 17 is installed at the bottom of the other adjustment structure 19 through a second mounting plate 18.

[0031] Two drive components 14 are respectively installed on the top of the arc-shaped detection chamber 7 near both ends, and drive gears 13 are installed on the output ends of both drive components 14.

[0032] The inner surfaces of the fixed plate 104 and the docking structure 103 are connected, the inner surface of the limiting structure 101 and the arc-shaped detection chamber 7 are connected, the external gear structure 102 and the drive gear 13 are meshed and connected, the two ends of the docking structure 103 and the two ends of the inner wall of the limiting structure 101 are connected by ball bearings, which can increase the stability of the rotation of the external gear structure 102. The adjustment structure 19 includes a bracket and an adjustment head. The second drive structure 16 provides rotational driving force for the adjustment head, which drives the mounting plate to adjust the angle and improve the detection range. The bottom of the mounting plate is equipped with an infrared sensor for auxiliary positioning. With the help of a camera, precise positioning can be achieved, so that the workpiece can stop at the corresponding position. The drive gear 13 passes through the through hole 15 and meshes with the external gear structure 102. The structure of the drive structure is the same as that of the drive assembly 14.

[0033] Both sides of the testing table 4 are equipped with feeding components 5. The feeding components 5 include a support frame 501 and a placement plate 502. The support frame 501 is used to install the placement plate 502 on both sides of the testing table 4.

[0034] The placement plate 502 is used to place the workpiece.

[0035] A control box 2 is installed on the top of the base plate 1. A door 3 is rotatably connected to the front of the control box 2. An operation screen 9 is installed on the front of the detection platform 4.

[0036] The door 3 is equipped with a lock. The control box 2 contains a power switch and a controller for controlling the operation of the equipment. The operation panel 9 allows manual control of the equipment operation and setting of operating parameters.

[0037] The top of the arc-shaped detection chamber 7 is equipped with a protective component 8, which includes a magnetic frame 801 and a protective frame 802. The magnetic frame 801 is used to install the protective frame 802 on the top of the arc-shaped detection chamber 7.

[0038] The protective frame 802 provides protection for the drive component 14.

[0039] The drive assembly 14 includes a protective shell 141, a drive component 142, and an angle sensor 143. The protective shell 141 is used to mount the drive component 142, which provides rotational driving force.

[0040] The drive component 142 has a self-locking function, and the angle sensor 143, together with the controller of the drive component 142, can accurately control the rotation speed and rotation angle.

[0041] The conveying assembly 6 includes a first drive structure 601, a rotating shaft 602, and a conveyor belt 603. The rotating shaft 602 is used to install the conveyor belt 603 inside the arc-shaped detection chamber 7.

[0042] The first drive structure 601 provides rotational power to one of the rotating shafts 602.

[0043] The working principle of the AI-based automated visual inspection device provided by this utility model is as follows:

[0044] Two adjustment components 10 are installed inside the arc-shaped detection chamber 7. Then, a small-field-of-view high-magnification lens 17 is mounted on the adjustment structure 19 of the adjustment component 10 via the second mounting plate 18, while a large-field-of-view low-magnification lens 11 is mounted on the adjustment structure 19 of the other adjustment component 10 via the first mounting plate 12. The drive component 14 drives the external gear structure 102 to rotate, thereby adjusting the angles of the large-field-of-view low-magnification lens 11 and the small-field-of-view high-magnification lens 17. The second drive structure 16 and the adjustment structure 19 then drive the large-field-of-view low-magnification lens 11 and the small-field-of-view high-magnification lens 17 to adjust their angles left and right, increasing the detection area. In actual use, the workpiece is slowly moved inside the arc-shaped detection chamber 7 via the conveying component 6. During this process, the first mounting... The infrared sensors at the bottom of the tray 12 and the second mounting tray 18 send a signal after the workpiece moves to the corresponding position, causing the conveying assembly 6 to stop working. Then, the adjustment assembly 10, the second drive structure 16, and the adjustment structure 19 are activated to drive the large field-of-view low-magnification lens 11 to quickly scan the entire workpiece. The AI ​​algorithm uses the image from this channel to locate the workpiece position, determine whether the overall outline is qualified, and mark suspicious defect areas. After completing the preliminary inspection, the conveying assembly 6 is activated to move the workpiece to below the small field-of-view high-magnification lens 17. With the help of the infrared sensor, the workpiece is stopped at the corresponding position. Then, the adjustment assembly 10, the second drive structure 16, and the adjustment structure 19 drive the small field-of-view high-magnification lens 17 to move precisely to the suspicious area under the guidance of the AI ​​algorithm, and take high-definition detailed pictures of the workpiece to further identify minor defects.

[0045] Compared with related technologies, the automated visual inspection device based on AI algorithms provided by this utility model has the following beneficial effects:

[0046] To improve the accuracy of the AI-based automated visual inspection device, two adjustment components 10 are installed inside the arc-shaped inspection chamber 7. Then, a small-field-of-view high-magnification lens 17 is mounted on the adjustment structure 19 of the adjustment component 10 via a second mounting plate 18, while a large-field-of-view low-magnification lens 11 is mounted on the adjustment structure 19 of the other adjustment component 10 via a first mounting plate 12. By driving the external gear structure 102 through the drive component 14, the angles of the large-field-of-view low-magnification lens 11 and the small-field-of-view high-magnification lens 17 can be adjusted. The second drive structure 16 and the adjustment structure 19 drive the large field-of-view low-magnification lens 11 and the small field-of-view high-magnification lens 17 to adjust their angles left and right, thereby increasing the detection area. This design uses the large field-of-view low-magnification lens 11 to ensure detection efficiency and the small field-of-view high-magnification lens 17 to ensure detection accuracy, realizing a collaborative working mode of "rapid screening and accurate verification". It reduces unnecessary high-magnification shooting areas, reduces data processing volume, and improves the computing efficiency of AI algorithms. It is suitable for scenarios that require both rapid batch detection and identification of minute defects, such as mobile phone screens and precision molds.

[0047] The above description is merely an embodiment of this utility model and does not limit the patent scope of this utility model. Any equivalent structural or procedural transformations made based on the content of this utility model specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this utility model.

Claims

1. An automated visual inspection device based on AI algorithms, characterized in that, include: Base plate; A testing platform is fixedly connected to the top of a base plate. An arc-shaped testing chamber is installed on the top of the testing platform. A conveying assembly is installed at the bottom of the inner wall of the arc-shaped testing chamber. Adjustment assemblies are installed near both sides inside the arc-shaped testing chamber. Each adjustment assembly includes a limiting structure, an external gear structure, a docking structure, and a fixing plate. The docking structure is used to rotatably connect the external gear structure to the inside of the limiting structure. Adjustment structures are installed on the inner surfaces of the two adjustment assemblies. A second drive structure is installed at one end of each adjustment structure. A first mounting plate is installed at the bottom of one adjustment structure, and a large field-of-view low-magnification lens is installed at the bottom of the first mounting plate. A small field-of-view high-magnification lens is installed at the bottom of the other adjustment structure via a second mounting plate. Two drive components are installed on the top of the arc-shaped detection chamber near both ends, and drive gears are installed at the output ends of both drive components.

2. The automated visual inspection device based on AI algorithm according to claim 1, characterized in that, Both sides of the testing station are equipped with feeding assemblies, which include support frames and placement plates. The support frames are used to install the placement plates on both sides of the testing station.

3. The automated visual inspection device based on AI algorithm according to claim 1, characterized in that, A control box is mounted on the top of the base plate, and a door is rotatably connected to the front of the control box. An operation screen is mounted on the front of the testing platform.

4. The automated visual inspection device based on AI algorithm according to claim 1, characterized in that, A protective assembly is installed on the top of the arc-shaped detection chamber. The protective assembly includes a magnetic frame and a protective frame. The magnetic frame is used to install the protective frame on the top of the arc-shaped detection chamber.

5. The automated visual inspection device based on AI algorithm according to claim 1, characterized in that, The drive assembly includes a protective housing, a drive component, and an angle sensor. The protective housing is used to mount the drive component that provides rotational driving force.

6. The automated visual inspection device based on AI algorithm according to claim 1, characterized in that, The conveying assembly includes a first drive structure, a rotating shaft, and a conveyor belt, the rotating shaft being used to install the conveyor belt inside the arc-shaped detection chamber.

Images