Equipment and method for detecting laser carving appearance of oil cup

Abstract

The invention relates to the technical field of appearance detection, in particular to oil cup laser carving appearance detection equipment and a detection method. The equipment comprises a grabbing mechanism, a deep learning vision assembly and a controller. The controller is connected with the grabbing mechanism and the deep learning vision assembly, the grabbing mechanism is used for receiving control instructions of the controller, the control instructions comprise a grabbing instruction and a rotating instruction, a to-be-detected oil cup is grabbed from a feeding position according to the grabbing instruction, and the rotating instruction is used for rotating the to-be-detected oil cup. The oil cup to be detected is rotated according to the rotation instruction; the deep learning vision assembly is used for carrying out laser etching appearance detection on the oil cup to be detected according to a preset deep learning algorithm and sending a detection result to the controller; the controller is used for sending a corresponding control instruction to the grabbing mechanism according to the detection result. The detection equipment provided by the invention can improve the precision of laser carving appearance detection of the oil cup.

Description

Technical Field

[0001] This invention relates to the field of appearance inspection technology, and in particular to an oil cup laser engraving appearance inspection device and inspection method. Background Technology

[0002] An electronic cigarette is an electronic product that mimics a traditional cigarette, mainly consisting of three parts: a cartridge containing nicotine solution, an atomizer, and a battery. The cartridge, also known as the e-liquid cup, is cylindrical in shape and has at least one inlet or outlet for nicotine.

[0003] During the production of e-cigarette cups, laser engraving machines are used to engrave the surface of the e-cigarette cups to improve their performance (e.g., making them scratch-resistant and non-slip) or for aesthetic purposes. After laser engraving, the engraved content is usually inspected manually. However, manual inspection is time-consuming, labor-intensive, and prone to missed or incorrect inspections.

[0004] Regarding the aforementioned problems with the appearance inspection of oil cups, Chinese patent application CN208432552U discloses an oil cup inspection device, including a worktable for placing oil cups and several inspection lamps arranged side by side on the worktable; the inspection lamps include a tubular support column vertically fixed to the worktable, a turntable disposed at the top of the support column, and a rotating shaft sleeved inside the support column and fixed at its top to the turntable; the rotating shaft is driven to rotate by a transmission mechanism disposed below the worktable, thereby causing the turntable to rotate at the top of the support column; several lighting lamps are disposed vertically on the outside of the support column; the oil cup is sleeved on the turntable and support column from its opening and is driven to rotate by the turntable.

[0005] However, this device can only be used for the appearance inspection of oil cups made of transparent or semi-transparent materials. It cannot be used to inspect oil cups made of opaque materials, and manual judgment is required based on the appearance of the projected light to determine whether they are qualified. There is still a possibility of missed detection or false detection. Summary of the Invention

[0006] To address the aforementioned technical problems of false detection and missed detection when inspecting the appearance of oil cups, the present invention provides solutions in the following aspects.

[0007] In a first aspect, the present invention provides an oil cup laser engraving appearance inspection device, including a gripping mechanism, a deep learning vision component, and a controller; the controller is connected to the gripping mechanism and the deep learning vision component, wherein the gripping mechanism is used to receive control commands from the controller, the control commands including gripping commands and rotation commands, and gripping the oil cup to be inspected from the loading position according to the gripping commands, and rotating the oil cup to be inspected according to the rotation commands; the deep learning vision component is used to perform laser engraving appearance inspection on the oil cup to be inspected according to a preset deep learning algorithm, and to send the inspection results to the controller; the controller is used to send corresponding control commands to the gripping mechanism according to the inspection results.

[0008] Beneficial effects: By gripping the oil cup to be inspected and rotating it during inspection, the deep learning vision component can perform 360° inspection of the laser-engraved appearance of the oil cup according to a preset deep learning algorithm, thus ensuring the completeness and accuracy of the inspection. Furthermore, the inspection process requires no manual intervention; the deep learning vision component completes the inspection, avoiding the problems of missed detections and false detections that exist in manual inspection, thereby improving inspection efficiency.

[0009] Furthermore, the device also includes a conveying assembly for conveying the oil cup to be tested to the loading position.

[0010] Beneficial effect: By conveying the oil cup to be tested through the conveying component, manual conveying is eliminated, further improving the efficiency of testing.

[0011] Furthermore, the conveying assembly includes a conveyor belt and a transport component, the transport component being disposed on the conveyor belt for carrying a tray containing multiple oil cups to be tested.

[0012] Furthermore, the gripping mechanism includes a gripping module located on the opposite side of the deep learning vision component; the gripping module includes multiple grippers, each gripper having a rotating component connected to its top end; the grippers are used to grip an oil cup to be inspected, and the rotating component is used to rotate the grippers to rotate the oil cup to be inspected gripped by the grippers.

[0013] Furthermore, the deep learning vision component includes multiple deep learning cameras, the lenses of which are oriented toward the gripper in the grasping module.

[0014] Furthermore, the gripping mechanism also includes a moving unit connected to the gripping module for moving the gripping module.

[0015] The moving unit includes a horizontal moving component and a vertical moving component, the horizontal moving component being connected to the vertical moving component, and the vertical moving component also being connected to the grasping module; wherein, the horizontal moving component is used to move the grasping module horizontally and assist the deep learning camera in the deep learning vision component to complete focusing; the vertical moving component is used to raise and lower the grasping module.

[0016] Furthermore, the device also includes a light source, which is disposed below the gripping module.

[0017] Beneficial effects: By setting a light source at the bottom of the gripping module, the interference of brightness on imaging can be avoided and the clarity of the image can be improved when performing visual inspection on the oil cup to be inspected, thereby improving the accuracy of the inspection.

[0018] In a second aspect, the present invention provides a method for laser engraving appearance inspection of oil cups, applied to the oil cup laser engraving appearance inspection equipment described in the first aspect. The method includes: in response to an oil cup to be inspected reaching a loading position, sending a gripping command to a gripping mechanism to cause the gripping mechanism to grip the oil cup to be inspected from the loading position; sending a flying image command to a deep learning vision component and a rotation command to the gripping mechanism to cause the deep learning vision component to perform laser engraving appearance inspection on the oil cup to be inspected according to a preset deep learning algorithm; in response to the completion of inspection, obtaining the inspection result of the oil cup to be inspected, the inspection result including qualified and unqualified; if the inspection result is unqualified, sending a discard command to the gripping mechanism; if the inspection result is qualified, sending a transfer command to the gripping mechanism.

[0019] Beneficial effects: Compared to manual inspection, inspection using deep learning vision components can quickly and accurately determine whether the laser engraving content on the oil cup to be inspected is qualified.

[0020] Furthermore, the deep learning algorithm is a convolutional neural network model or a generative adversarial network model.

[0021] The beneficial effects of the present invention are as follows: Compared with the prior art, the device of the present invention automatically acquires and detects the laser-engraved appearance of the oil cup through a deep learning vision component, without the need for manual observation of the laser-engraved content or the light projection emitted by the light source inside the oil cup, thereby avoiding the problems of false detection and missed detection caused by human factors, and thus improving the accuracy of detection. Attached Figure Description

[0022] Figure 1 This is a schematic diagram illustrating the overall structure of an oil cup laser engraving appearance inspection device according to an embodiment of the present invention;

[0023] Figure 2This is a schematic enlarged view of a partial enlarged view of an oil cup laser engraving appearance inspection device according to an embodiment of the present invention;

[0024] Figure 3 This is a flowchart illustrating an oil cup laser engraving appearance inspection method according to an embodiment of the present invention.

[0025] Figure Descriptions: 1-Gripping mechanism; 2-Deep learning vision component; 3-Conveying component; 4-Light source; 5-Machinery; 6-Laser engraving machine; 11-Gripping module; 12-Moving unit; 21-Deep learning camera; 22-Support frame; 23-Mounting plate; 31-Conveyor belt; 32-Transporting component; 110-Grip clamp; 120-Rotating component; 130-Transmission belt; 140-Motor; 150-Solenoid valve; 160-Air pipe; 1201-Horizontal moving component; 1202-Vertical moving component. Detailed Implementation

[0026] 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, not all, of the embodiments of the present invention. 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.

[0027] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0028] Figure 1 This is a schematic diagram illustrating the overall structure of an oil cup laser engraving appearance inspection device according to an embodiment of the present invention.

[0029] To address the technical problems of missed detections and false detections in the laser engraving appearance inspection of oil cups in existing technologies, in a first aspect, the present invention provides an oil cup laser engraving appearance inspection device. For example... Figure 1 As shown, the device of the present invention includes: a gripping mechanism 1, a deep learning vision component 2, a conveying component 3, a light source 4, a machine base 5, a laser engraving machine 6, and a controller (not shown in the figure).

[0030] In this embodiment, the machine base 5 is used to support and fix the gripping mechanism 1, the deep learning vision component 2, the conveying component 3, the light source 4, the laser engraving machine 6, and the controller.

[0031] In one embodiment, the controller is a PLC (Programmable Logic Controller), which is connected to the gripping mechanism 1, the deep learning vision component 2, the light source 4, etc., and is used to send control commands, receive the detection results of the oil cup to be detected, and send corresponding control commands to the gripping mechanism 1 according to the detection results.

[0032] The gripping mechanism 1 is used to receive control commands from the controller, and according to the corresponding control commands, grip the oil cup to be tested from the loading position to the testing area, and rotate the oil cup to be tested during testing. In this embodiment, the control commands include gripping commands and rotation commands. The gripping mechanism 1 grips the oil cup to be tested according to the gripping commands and rotates the oil cup to be tested according to the rotation commands.

[0033] In one embodiment, the grasping mechanism 1 includes a grasping module 11, such as Figure 2 As shown, the gripping module 11 includes multiple grippers 110 and multiple rotating components 120, with a rotating component 120 connected to the top of each gripper 110. The grippers 110 are used to grip the oil cup to be inspected, and the rotating components 120 are used to rotate the grippers 110 connected to them, thereby rotating the oil cup to be inspected. This allows the deep learning camera 21 to accurately and completely capture the entire laser engraving content of the oil cup to be inspected.

[0034] In this embodiment, there are 12 grippers 110, which can grip 12 oil cups to be inspected. By setting multiple grippers 110, multiple oil cups to be inspected can be inspected at the same time, thereby improving the efficiency of laser engraving appearance inspection.

[0035] In addition, the gripping module 11 also includes a drive belt 130 and a motor 140. The drive belt 130 is connected to each rotating component 120, and the motor 140 is connected to the controller. When the motor 140 receives a rotation command from the controller, it controls the drive belt 130 to rotate at a specified speed. When the drive belt 130 rotates, it drives the rotating component 120 to rotate, thereby realizing the rotation of the gripper 110, and thus controlling the oil cup to be tested to rotate by a specified angle, such as 360°, during testing.

[0036] By rotating the oil cup to be inspected once during inspection, the deep learning vision component 2 can accurately and completely capture the laser engraving content of the oil cup to be inspected, thereby avoiding the problems of missed detection and false detection.

[0037] In this embodiment, the gripping module 11 also includes multiple solenoid valves 150 and multiple air pipes 160. A pair of solenoid valves 150 controls the flow of gas in one air pipe 160. Each air pipe 160 passes through the interior of the rotating component 120 and connects to the corresponding gripper 110. Therefore, the opening and closing of the gripper 110 can be controlled by a cylinder to grip the oil cup to be tested. Specifically, when gripping is required, the solenoid valve 150 controlling the clamping will open. At this time, compressed gas will enter the clamping chamber of the cylinder (not shown in the figure) through the air pipe 160, pushing the piston (not shown in the figure) to close the gripper 110 and grip. When release is required, the solenoid valve 150 controlling the release on the other side will open. At this time, compressed air enters the release chamber of the cylinder, pushing the piston to move in the opposite direction, and the gripper 110 opens and releases.

[0038] In one embodiment, the deep learning vision component 2 includes multiple deep learning cameras 21, which are used to acquire images of the oil cup to be detected in the detection area. Each deep learning camera 21 integrates a preset deep learning algorithm and edge computing capabilities, enabling real-time analysis, learning, and decision-making based on the images of the oil cup to be detected. In this embodiment, the deep learning algorithm used is a convolutional neural network model or a generative adversarial network model. Compared to manual detection of intricate laser engravings, the false positive and false negative rates of detection using a convolutional neural network model or a generative adversarial network model are lower, thereby improving detection accuracy.

[0039] In this embodiment, there are six deep learning cameras 21, each capable of inspecting two oil cups simultaneously. By having one deep learning camera 21 inspect two oil cups at the same time, both inspection accuracy and efficiency are ensured. In other optional embodiments, those skilled in the art can adjust the number of deep learning cameras 21 according to actual needs. For example, only one camera can be used, and the oil cups can be inspected sequentially to reduce inspection costs; or the number of deep learning cameras 21 can be increased to improve the efficiency of laser engraving appearance inspection of oil cups.

[0040] In addition, the deep learning vision component 2 also includes a support frame 22 for supporting the deep learning camera 21. The top of the support frame 22 is provided with a mounting plate 23, which has multiple mounting holes and grooves for fixing the deep learning camera 21. The bottom of the deep learning camera 21 has multiple inserts that mate with the mounting holes and sliding rods that mate with the grooves. In practical applications, operators can adjust the position of the deep learning camera 21 on the mounting plate 23 according to actual needs through the mounting holes.

[0041] In this embodiment, the lens of the deep learning camera 21 is directed toward the gripper 110 in the gripping module 11, that is, the deep learning camera 21 and the gripper 110 are opposite each other, and the gripping module 11 is located on the opposite side of the deep learning camera 21.

[0042] By setting up the support frame 22, the lens of the deep learning camera 21 can be directly facing the oil cup to be tested to acquire images, thereby obtaining clear and complete imaging. By setting up the mounting plate 23, when a deep learning camera 21 malfunctions, the distance between the deep learning camera 21 and the oil cup to be tested needs to be adjusted, or a higher precision deep learning camera needs to be used, it can be easily and quickly replaced.

[0043] In one embodiment, the moving unit 12 is connected to the gripping module 11 to move the gripping module 11 as a whole. Specifically, the moving unit 12 includes a horizontal moving component 1201 and a vertical moving component 1202. One end of the vertical moving component 1202 is connected to the horizontal moving component 1201, and the other end is connected to the gripping module 11. When the vertical moving component 1202 moves vertically, it drives the gripping module 11 to move vertically as a whole, thereby realizing the vertical movement of the gripper 110. When the horizontal moving component 1201 moves horizontally, it drives the vertical moving component 1202 to move horizontally, thereby driving the gripping module 11 to move horizontally, and thus realizing the horizontal movement of the gripper 110. By setting the moving unit 12, the movement of the oil cup to be tested can be realized. In addition, the horizontal moving component 1201 can also assist the deep learning camera 21 in completing focusing, thereby enabling the acquisition of clear images and avoiding false detections or invalid detections caused by blurry images.

[0044] In other optional embodiments, only the horizontal moving component 1201 can be provided. This horizontal moving component 1201 is connected to the gripping module 11. During gripping, the gripper 110 moves horizontally to the loading position, grips the oil cup to be tested at the loading position, and then moves horizontally to the set testing point or testing area. Alternatively, only the vertical moving component 1202 can be provided. This vertical moving component 1202 is connected to the gripping module 11. During gripping, the gripper 110 is at a higher position at the loading position. When the oil cup to be tested reaches the loading position, the gripper 110 descends to grip the oil cup to be tested, and then rises to the set position (testing point or testing area). During testing, the oil cup to be tested is rotated. After testing, the oil cup to be tested is returned to its original position.

[0045] In one embodiment, the conveying assembly 3 is used to transport the oil cups to be inspected to the loading position for visual inspection. Specifically, the conveying assembly 3 includes a conveyor belt 31 and a transport component 32. The transport component 32 is disposed on the conveyor belt 31, the conveyor belt 31 is used to move the transport component 32, and the transport component 32 is used to carry a tray containing multiple oil cups to be inspected, thereby realizing the transport of the oil cups to be inspected.

[0046] In this embodiment, the tray may be provided with multiple receiving components. These receiving components are used to fix the oil cup to be tested, preventing it from falling or changing position during transport. Therefore, when the tray moves to the loading position, the gripping module 11 can accurately grip the oil cup to be tested. The receiving component can be set according to actual needs, for example, it can be set as a fixing member that is consistent with the outer contour of the oil cup, but slightly larger than the outer contour of the oil cup, and the oil cup to be tested is fixed by insertion. In other optional embodiments, the transport component 32 and the tray can also be integrally formed.

[0047] It should be noted that only one section of conveyor belt is used when transporting the oil cup to be tested.

[0048] In an optional embodiment, the conveying component 3 further includes a positioning device (e.g., GPS), which is installed on the transport component 32 to obtain the positioning information of the transport component 32 and send the positioning information to the controller. When the positioning information is the position of the loading position, the controller sends a gripping command to the gripping mechanism 1.

[0049] In this embodiment, as Figure 2 As shown, the light source 4 is installed below / at the bottom of the gripper 110 in the gripping module 11 to provide illumination. By placing the light source 4 at the bottom of the gripper 110, the images captured by the deep learning camera 21 are clearer, and the interference of brightness changes in the detection environment on the detection is reduced, further improving the detection accuracy.

[0050] In this embodiment, the laser engraving machine 6 is used to laser engrave the oil cups to obtain the oil cups to be inspected. After the laser engraving is completed, the robot loads multiple oil cups to be inspected onto a tray, and then places the tray onto the transport component 32. By placing the oil cups to be inspected onto the transport component 32 in a whole-tray manner, the inspection efficiency can be improved.

[0051] The entire testing process requires no manual inspection, thus avoiding the problems of missed or false detections that exist with manual inspection, while also improving testing efficiency.

[0052] Figure 3 This is a flowchart illustrating an oil cup laser engraving appearance inspection method according to an embodiment of the present invention.

[0053] In a second aspect, the present invention provides a method for laser-engraved appearance inspection of oil cups, applied to the laser-engraved appearance inspection equipment for oil cups described in the first aspect, employing a centralized controller to drive, control, and transmit information to various electronic / electrical mechanisms. For example... Figure 3 As shown, the method of the present invention includes the following steps.

[0054] S101. In response to the oil cup to be tested reaching the loading position, a gripping command is sent to the gripping mechanism so that the gripping mechanism can grip the oil cup to be tested from the loading position.

[0055] Specifically, the robot, following the controller's instructions, loads the laser-engraved oil cups into a tray, places the tray onto a transport component, and then transports the oil cups to the loading position via a conveyor belt. When the oil cup is detected at the loading position (for example, by obtaining the positioning information of the transport component through a positioning device, and then the controller determines whether the oil cup has reached the loading position based on whether the positioning information matches the position of the loading position), a gripping command is sent to the gripping mechanism, enabling the gripping mechanism to grab the oil cup from the loading position and transport it to the inspection area via a moving unit.

[0056] S102, Send a flying camera command to the deep learning vision component and a rotation command to the gripping mechanism, so that the deep learning vision component performs laser engraving appearance inspection on the oil cup to be inspected according to the preset deep learning algorithm.

[0057] In this embodiment, the deep learning algorithm employs a pre-trained convolutional neural network model or generative adversarial network model for laser engraving appearance detection.

[0058] In one embodiment, the method for obtaining a convolutional neural network model for laser engraving appearance detection includes: feeding multiple standard template images into the convolutional neural network model for deep learning and training, so that it has the basic judgment ability of laser engraving appearance defects of oil cups, and obtaining a first edge feature matrix.

[0059] High-speed image capture (HSV) technology is a technique in industrial vision used to achieve efficient and accurate image acquisition and processing in high-speed motion scenarios. While a deep learning camera captures images, a gripping mechanism rotates the gripped oil cup to be inspected 360°, thus obtaining an HSV image of the entire laser engraving content on the oil cup. This HSV image is then input into a convolutional neural network model for laser engraving appearance inspection to obtain a second edge feature matrix. The cosine similarity between the first and second edge feature matrices is calculated, and this cosine similarity is used as the matching degree. If the matching degree is less than a set threshold (e.g., 98%), the laser engraving content of the oil cup is deemed unqualified (i.e., the detection result is unqualified). If the matching degree is greater than or equal to the set threshold, the laser engraving content of the oil cup is deemed qualified (i.e., the detection result is qualified).

[0060] S103. In response to the completion of the detection, obtain the detection result of the laser engraving appearance, and send the corresponding control command to the gripping mechanism according to the detection result.

[0061] Once the oil cup to be inspected has rotated 360°, the image inspection is complete. At this point, the deep learning vision component sends the inspection result to the controller, which then sends corresponding control commands to the gripping mechanism based on the result. Specifically, if the inspection result is unqualified, a discard command is sent to the gripping mechanism, causing it to place the unqualified product in the unqualified product storage area; if the inspection result is qualified, a transfer command is sent to the gripping mechanism, causing it to place the qualified product back into the pallet on the conveyor belt for transport to the next workstation.

[0062] Compared with existing technologies, the detection method of the present invention does not require manual inspection of laser engraving content. The deep learning vision component performs 360° all-round photo inspection of the oil cup to be inspected according to the preset deep learning algorithm, thereby avoiding the risk of false detection and missed detection caused by human factors, while improving detection efficiency. It can also be applied to the laser engraving appearance inspection of oil cups made of opaque materials.

[0063] In the description of this specification, "multiple" means at least two, such as two, three or more, unless otherwise explicitly specified. Furthermore, the steps described above are for clarity only; in implementation, they can be combined into one step or some steps can be broken down into multiple steps, as long as they include the same logical relationships.

[0064] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.

Claims

1. A laser engraving appearance inspection device for oil cups, characterized in that, It includes a grasping mechanism, a deep learning vision component, and a controller; the controller is connected to the grasping mechanism and the deep learning vision component, wherein, The gripping mechanism is used to receive control commands from the controller, the control commands including gripping commands and rotation commands, and to grip the oil cup to be tested from the loading position according to the gripping commands, and to rotate the oil cup to be tested according to the rotation commands. The deep learning vision component is used to perform laser engraving appearance detection on the oil cup to be detected according to a preset deep learning algorithm, and to send the detection results to the controller. The controller is used to send corresponding control commands to the grasping mechanism based on the detection results.

2. The oil cup laser engraving appearance inspection equipment according to claim 1, characterized in that, It also includes a conveying assembly for conveying the oil cup to be tested to the loading position.

3. The oil cup laser engraving appearance inspection equipment according to claim 2, characterized in that, The conveying assembly includes a conveyor belt and a transport component, the transport component being disposed on the conveyor belt for carrying a tray containing multiple oil cups to be tested.

4. The oil cup laser engraving appearance inspection equipment according to claim 1, characterized in that, The gripping mechanism includes a gripping module located on the opposite side of the deep learning vision component; the gripping module includes multiple grippers, each gripper having a rotating component connected to its top; the grippers are used to grip an oil cup to be inspected, and the rotating component is used to rotate the grippers to rotate the oil cup to be inspected gripped by the grippers.

5. The oil cup laser engraving appearance inspection equipment according to claim 4, characterized in that, The deep learning vision component includes multiple deep learning cameras, the lenses of which are oriented toward the gripper in the grasping module.

6. The oil cup laser engraving appearance inspection equipment according to claim 5, characterized in that, The gripping mechanism further includes a moving unit connected to the gripping module for moving the gripping module.

7. The oil cup laser engraving appearance inspection equipment according to claim 6, characterized in that, The moving unit includes a horizontal moving component and a vertical moving component, the horizontal moving component being connected to the vertical moving component, and the vertical moving component also being connected to the grasping module; wherein, the horizontal moving component is used to move the grasping module horizontally and assist the deep learning camera in the deep learning vision component to complete focusing; the vertical moving component is used to raise and lower the grasping module.

8. The oil cup laser engraving appearance inspection equipment according to claim 4, characterized in that, It also includes a light source, which is located below the gripping module.

9. A method for inspecting the appearance of an oil cup using laser engraving, characterized in that, The method, applied to the oil cup laser engraving appearance inspection equipment according to any one of claims 1-8, comprises: In response to the oil cup to be tested reaching the loading position, a gripping command is sent to the gripping mechanism so that the gripping mechanism can grip the oil cup to be tested from the loading position; Send a flying camera command to the deep learning vision component and a rotation command to the gripping mechanism so that the deep learning vision component can perform laser engraving appearance inspection on the oil cup to be inspected according to the preset deep learning algorithm. In response to the completion of the test, the test result of the oil cup to be tested is obtained, and the test result includes qualified and unqualified. If the detection result is unqualified, a discard command is sent to the gripping mechanism; if the detection result is qualified, a transfer command is sent to the gripping mechanism.

10. The method for inspecting the appearance of an oil cup using laser engraving according to claim 9, characterized in that, The deep learning algorithm is a convolutional neural network model or a generative adversarial network model.

Images