Intelligent detection system and method for laser carving appearance of oil cup

Abstract

The invention relates to the technical field of industrial automatic detection, in particular to an intelligent detection system and method for the laser carving appearance of an oil cup. The system comprises a conveying unit, an operation execution unit, an image acquisition unit, a material management unit and a control unit. The image acquisition unit adopts a multi-mode assembly to synchronously acquire texture images and three-dimensional shape information of the surface of the oil cup. The control unit comprises an image processing module, a motion control module and a process feedback module, and the image processing module performs fusion judgment based on multi-modal information; the motion control module controls an execution mechanism to complete rejecting of unqualified products and automatic feeding of qualified products according to a judgment result, so that a closed-loop material flow is formed; and the process feedback module analyzes the defect trend and sends a process adjustment instruction to the front laser etching machine. According to the invention, comprehensive detection of laser etching content and depth is realized, a full-process automatic closed loop of'detection-rejection-material supplement-feedback 'is constructed, and detection completeness, production line continuity and process stability are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation detection, in particular to an oil cup laser etching appearance intelligent detection system and method. BACKGROUND

[0002] In the production process of an oil cup (for example, a tobacco tar storage cup for a fogging device such as an electronic cigarette), in order to meet the product identification, anti-counterfeiting traceability or aesthetic requirements, laser etching is usually performed on the outer surface thereof to form specific characters, patterns or codes.

[0003] At present, there are mainly two ways to detect the quality of oil cup laser etching: one is the traditional manual visual sampling inspection, which is low in efficiency, high in labor intensity, and susceptible to the subjective state of the inspector, and has the risk of missed detection and misjudgment, and is difficult to meet the requirements of full inspection and consistency of modern high-speed production lines. The other is to use automatic visual detection equipment for online detection, such as the oil cup defect detection method and system based on deep learning disclosed in Chinese Patent No. CN121095182A, which improves the detection accuracy of defects in the oil cup body material (such as bubbles, cracks, dirty spots and scratches) by improving the image preprocessing algorithm (such as multi-frame alignment and noise reduction).

[0004] However, when such automatic visual detection scheme mainly targeting material defects is directly applied to the specific and multi-dimensional detection scenario of oil cup laser etching quality, its inherent limitations are exposed, and it is difficult to constitute a complete and effective solution: 1. Single detection dimension, unable to comprehensively evaluate the laser etching quality: most of the existing solutions rely only on 2D texture images for analysis, mainly for identifying the correctness, absence or dirtiness of the laser etching content. However, for the laser etching process, the depth consistency of the engraving (such as too shallow engraving leading to unclearness, and too deep engraving possibly causing material damage) is a very critical quality indicator. Only 2D images cannot effectively and accurately obtain and judge this three-dimensional topographic information, resulting in incomplete detection and difficulty in preventing depth defects from flowing into subsequent links.

[0005] 2. Incomplete system function, unable to form a production closed loop: most of the existing devices are one-way processes of "detection-screening". When unqualified products are detected and removed, a vacancy will be generated on the production line. If qualified products cannot be automatically supplemented in time, it will lead to chaos in the subsequent packaging or assembly process, either stopping the machine for manual feeding, affecting the overall production efficiency, or running with vacancies, disrupting the production rhythm. SUMMARY

[0006] The present application provides an oil cup laser etching appearance intelligent detection system and method to solve the technical problems of single detection dimension and incomplete system function in the prior art.

[0007] On one hand, the present invention provides an intelligent inspection system for the laser-engraved appearance of oil cups, used for inspecting and sorting oil cups after laser engraving. The inspection system includes: A conveying unit is used to carry and transport a fixture containing multiple oil cups; An operation execution unit is correspondingly set at the detection station of the conveying unit, used to pick up the oil cup from the conveying unit and move the oil cup to a detection position that is separated from the conveying unit; An image acquisition unit is set at the detection position. The image acquisition unit is a multimodal image acquisition component, used to simultaneously acquire the texture image and three-dimensional shape information of the oil cup located at the image acquisition station. The material management unit includes a collection module for receiving defective oil cups and a replenishment module for providing qualified oil cups; The control unit is communicatively connected to the conveying unit, image acquisition unit, operation execution unit, and material management unit, respectively. The control unit includes an image processing module, a motion control module, and a process feedback module. The image processing module is connected to the image acquisition unit and is used to acquire the texture image and three-dimensional morphology information, and to comprehensively judge the laser engraving quality on the surface of the oil cup based on preset judgment rules, and output the judgment result. The motion control module is connected to the image processing module, the operation execution unit, and the material management unit, and is used to control the operation execution unit and the material management unit to remove unqualified oil cups and replenish qualified oil cups according to the judgment result, so as to form a closed-loop material flow. The process feedback module is connected to the image processing module and is used to generate process parameter adjustment instructions for the preceding laser engraving equipment based on the laser engraving quality information in the judgment result.

[0008] This detection system simultaneously acquires two-dimensional texture images and three-dimensional morphological information of the oil cup surface, and comprehensively judges the laser quality based on preset judgment rules. It can not only identify appearance defects such as missing characters and blurred patterns, but also accurately quantify key three-dimensional process parameters such as laser engraving depth, edge steepness, and contour consistency. It can effectively distinguish between real laser engraving defects and surface interference (such as dust, water stains, reflections, etc.), and significantly reduce the false judgment rate and missed detection rate.

[0009] Simultaneously, an integrated closed-loop process of "inspection-sorting-replenishment-feedback" has been constructed. While rejecting defective products, the system automatically grabs qualified oil cups from the replenishment module to fill the empty spaces in the fixture, ensuring that the fixture is always fully loaded and flowing, avoiding interruptions in downstream assembly or packaging processes due to material shortages, and improving production line continuity and overall equipment efficiency. The process feedback module can dynamically generate compensation instructions for laser engraving equipment (such as galvanometer offset, laser power, scanning speed, etc.) based on the defect types and quantitative trends in historical judgment results, realizing a feedforward-feedback collaborative mechanism of "inspection as control", nipping quality problems in the bud and effectively reducing the risk of batch defects.

[0010] As a preferred embodiment of the present invention, the multimodal image acquisition component includes at least one industrial camera and a structured light projector; the structured light projector is used to project a light pattern onto the surface of the oil cup, the industrial camera is used to acquire an image of the oil cup surface modulated by the light pattern, and the image processing module obtains the three-dimensional morphology information by analyzing the image acquired by the industrial camera.

[0011] As a preferred embodiment of the present invention, the operation execution unit includes a first moving module and a second moving module symmetrically arranged on both sides of the detection station of the conveying unit; each moving module includes a clamping mechanism and a rotary drive mechanism for driving the clamping mechanism to rotate.

[0012] As a preferred embodiment of the present invention, the rotary drive mechanism includes a rotary motor and a belt drive mechanism driven by the rotary motor; the clamping mechanism includes a plurality of grippers connected to the belt drive mechanism, and the rotary motor drives all grippers to rotate synchronously through the belt drive mechanism.

[0013] As a preferred embodiment of the present invention, the collection module is a defective product collection box, and the replenishment module is a replenishment bin; the motion control module controls the operation execution unit to put the defective oil cup into the defective product collection box, and to pick up the qualified oil cup from the replenishment bin to replenish the empty space of the conveying unit.

[0014] As a preferred embodiment of the present invention, the image processing module is connected to a sample database and a model update unit; the image processing module stores the sample data whose judgment results are marked as suspicious or require verification and their verification results into the sample database; the model update unit calls the data in the sample database during non-production periods to retrain the recognition model used for comprehensive judgment in the image processing module in order to update the model parameters.

[0015] As a preferred embodiment of the present invention, the comprehensive determination of the laser engraving quality on the surface of the oil cup by the image processing module includes the following steps: Analyze the texture image to identify the laser-engraved characters and patterns, and obtain the first sub-result; Analyze the three-dimensional topography information, extract the depth distribution characteristics of the laser-engraved area, and determine whether the engraving depth is qualified according to the preset depth standard to obtain the second sub-result; Based on the first sub-result and the second sub-result, perform a logical AND operation to output the final laser engraving quality judgment result; The final judgment result is qualified only if the content of the first sub-result is correctly identified and the depth judgment of the second sub-result is qualified.

[0016] As a preferred embodiment of the present invention, when the determination results of multiple consecutive oil cups output by the image processing module all indicate that the laser engraving position has an offset in the same direction or the engraving depth has a deviation in the same trend, the process feedback module generates a control command including a position compensation amount or a laser power adjustment amount and sends it to the controller of the front-end laser engraving equipment.

[0017] As a preferred embodiment of the present invention, the control unit further includes a data statistics module, which is used to record the location, time or batch information of the occurrence of defective oil cups, and generate production early warning signals or send instructions to the motion control module to adjust the feeding strategy based on historical data statistical patterns.

[0018] On the other hand, the present invention also provides an intelligent detection method for the laser engraving appearance of an oil cup, applied to the detection system described above. The detection method includes the following steps: The conveying unit transports the fixture containing the oil cup to the testing station; The operation execution unit picks up the oil cup from the conveying unit and moves it to the detection position; The image acquisition unit simultaneously acquires the texture image and three-dimensional morphology information of the oil cup at the detection position; The image processing module comprehensively determines the laser engraving content, position, and engraving depth on the surface of the oil cup based on the texture image and three-dimensional morphology information, and outputs the determination result. Based on the determination result, the motion control module controls the operation execution unit and the material management unit to remove the unqualified oil cup and replace it with a qualified oil cup. The process feedback module generates an instruction to adjust the process parameters of the preceding laser engraving equipment based on the determination result.

[0019] The beneficial effects are: 1. This invention, through the coordinated control of the operation execution unit, material management unit, and motion control module, can simultaneously and automatically grab qualified products from the replenishment bin and accurately fill vacant workstations while rejecting defective products. This innovative mechanism completely changes the passive mode of traditional inspection stations that can only perform unidirectional "screening and rejection," ensuring that the fixtures at the output workstations are always fully loaded with qualified products. As a result, downstream processes such as packaging and assembly do not need to wait or adjust, achieving truly uninterrupted continuous flow production, improving overall production line efficiency, and completely eliminating downtime losses caused by material shortages.

[0020] 2. By statistically analyzing real-time detection data through the process feedback module, systematic process deviation trends can be intelligently identified, and precise process parameter adjustment instructions (such as galvanometer coordinate compensation and laser power correction values) can be automatically generated and fed back to the control system of the upstream laser engraving equipment in real time. This innovation upgrades the detection process from a passive "quality judge" to a proactive "process controller," realizing online monitoring and proactive correction of the production process. Through this feedforward control mechanism, the generation of batch defective products can be prevented from the source of the process, improving the first-pass yield and enabling the production process to have a self-stabilizing capability for continuous optimization.

[0021] 3. By employing a multimodal image acquisition unit that integrates two-dimensional texture imaging and three-dimensional topography measurement (such as structured light), the system can not only identify two-dimensional information such as characters and patterns engraved in laser engravings, but also accurately quantify the engraving depth and its uniformity. This overcomes the fatal shortcoming of existing technologies that rely solely on two-dimensional images, leading to missed detection of depth defects. For the first time, it achieves simultaneous and objective evaluation of the "shape" (content) and "depth" (quality) of laser engravings on the production line, elevating the detection dimension from two-dimensional to three-dimensional, and improving the completeness and accuracy of the detection. Attached Figure Description

[0022] Figure 1 A schematic diagram of a laser engraving appearance intelligent inspection system for oil cups; Figure 2 for Figure 1 A magnified view of part A in the image; Figure 3 This is a flowchart of an intelligent detection method for the laser engraving appearance of an oil cup.

[0023] Explanation of reference numerals in the attached figures: 11. Fixture; 21. First moving module; 22. Second moving module; 31. Industrial camera; 41. Defective product collection box; 42. Replenishment bin. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0025] In the description of this application, it should be noted that, unless otherwise stated, "multiple" means two or more. The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and are not limited in number; for example, a first object can be one or more.

[0026] The principles and essence of the present invention will be explained in detail below with reference to several representative embodiments.

[0027] Example 1: like Figure 1 and 2 The system shown is an intelligent inspection system for the appearance of laser engraving on oil cups, which is integrated into the laser engraving production line of oil cups and interacts and coordinates with the upstream laser engraving machine.

[0028] The detection system includes a conveying unit, an operation execution unit, an image acquisition unit, a material management unit, and a control unit. These five units are electrically connected and work together with a communication network to form a complete intelligent detection closed loop.

[0029] The conveying unit, serving as the material inlet and flow basis of the system, includes a linear module, a servo motor, and a dedicated fixture 11. The fixture 11 is equipped with positioning slots that conform to the shape of the oil cups, allowing it to hold multiple oil cups in an array (e.g., 10x2). The servo motor drives the linear module to precisely convey the fixture 11, fully loaded with oil cups, from the upstream connection point to the inspection station inside the system, and then sends it out after inspection.

[0030] The operation execution unit is responsible for the gripping, transferring, flipping, and placing of the oil cup. This unit includes a first moving module 21 and a second moving module 22 symmetrically arranged on both sides of the inspection station. The first moving module 21 and the second moving module 22 have identical structures. Each moving module (taking the first moving module 21 as an example) includes: The motion actuator consists of a horizontal linear motor and a vertical cylinder, enabling movement in the X and Z axes.

[0031] Rotary drive mechanism: includes a servo motor (rotary motor) and a synchronous belt drive mechanism driven by the motor.

[0032] Clamping mechanism: includes multiple (e.g., 10) grippers fixed to the driven shaft of the belt drive mechanism. When the rotary motor is activated, all grippers are driven to rotate synchronously via the belt drive mechanism.

[0033] Under the command of the control unit, the moving execution component drives the gripper to descend and pick up the oil cup on the fixture 11. After being lifted, it moves horizontally to the preset detection position (this position is away from the conveying unit to avoid interference). Then, the rotation drive mechanism can drive the oil cup to rotate a specific angle (such as 180°) for multi-face detection. Finally, according to the command, the oil cup is put back into the fixture 11 or transferred to the material management unit.

[0034] The image acquisition unit is positioned at the detection location to ensure that images of multiple surfaces of the oil cup can be acquired synchronously and with high precision to obtain complete information about the laser-engraved surfaces of the oil cup. The image acquisition unit is a multimodal image acquisition component, which includes at least one high-resolution industrial camera 31 and a structured light projector.

[0035] A structured light projector projects a pre-coded optical pattern (such as striped light or grid light) onto the surface of an oil cup located at the detection position. Preferably, the structured light projector uses a combination of coaxial light and low-angle diffused light to highlight the outline and contrast of the laser-engraved characters and patterns, while suppressing specular reflection interference from metal or high-gloss plastic surfaces.

[0036] The industrial camera 31 acquires images of distorted light patterns modulated by the surface morphology of the oil cup from a certain angle.

[0037] Using algorithms such as triangulation, the image processing module can analyze and calculate the three-dimensional coordinates of each point on the surface of the oil cup based on the distorted image, thereby obtaining high-precision three-dimensional topographic information (i.e., micron-level depth information). Simultaneously, the texture image acquired by the industrial camera 31 under uniform lighting is used to analyze the two-dimensional features of the laser-engraved characters and patterns. Thus, the system achieves simultaneous acquisition of texture and three-dimensional topographic information. This method exhibits excellent anti-interference capabilities on metal or highly reflective surfaces, overcoming the problem of signal loss in strongly reflective areas by traditional laser line scanning, ensuring stable acquisition of key three-dimensional features such as the depth and edge contour of the laser-engraved groove.

[0038] The industrial camera 31 and the structured light projector are triggered by the control unit to ensure that the two-dimensional texture image and the structured light modulated image are strictly synchronized in time and space, avoiding registration errors caused by oil cup micro-movement or rotation, providing a high-quality, pixel-level aligned data foundation for subsequent multimodal feature fusion, and improving the accuracy and repeatability of defect identification.

[0039] The material management unit includes a collection module and a replenishment module.

[0040] The collection module is a defective product collection box 41, which is located on one side of the rack and is used to store oil cups that are judged to be defective.

[0041] The feeding module is a feeding bin 42, located on the other side of the frame. It contains confirmed qualified oil cups, and usually uses a vibratory feeder or directional feed channel to ensure orderly output.

[0042] This unit works in conjunction with the operation execution unit to complete the material handling process of "removing defective materials and replenishing good materials" under the command of the control unit.

[0043] The control unit is integrated into the industrial control cabinet and adopts a distributed control architecture. The control unit includes a motion control module, an image processing module, and a process feedback module. Motion control module: Preferably implemented by a programmable logic controller (PLC). The motion control module is directly electrically connected to the interfaces of the servo motors of the conveying unit, the various motor cylinders of the operation execution unit, and the material management unit, and is responsible for the precise sequential control and logical interlocking of all underlying motion mechanisms.

[0044] The image processing module and process feedback module are integrated into an industrial computer (industrial control computer). The image processing module connects to the industrial camera 31 of the image acquisition unit via Gigabit Ethernet to receive and process image data. The process feedback module has standard industrial communication interfaces (such as Modbus TCP, Ethernet / IP).

[0045] Data Interaction: The industrial computer (carrier module) and PLC (motion control module) exchange data at high speed through industrial Ethernet to achieve real-time synchronization of instructions and status.

[0046] like Figure 3 As shown, the system's workflow and intelligent judgment method are as follows: S1: The conveying unit transports the fixture containing the oil cup to the inspection station.

[0047] This step is loading and positioning, specifically: the conveying unit transports the fixture carrying the laser-engraved oil cup to the inspection station and positions it precisely.

[0048] S2: The operation execution unit picks up the oil cup from the conveying unit and moves it to the detection position.

[0049] This step involves synchronous gripping and transfer. Specifically, the first and second moving modules move synchronously to grip the left and right rows of oil cups on the fixture and transfer them to the corresponding detection positions.

[0050] S3: The image acquisition unit simultaneously acquires the texture image and three-dimensional shape information of the oil cup at the detection position.

[0051] This step involves multimodal image acquisition, specifically: the image acquisition unit is activated, the structured light projector projects a light pattern, and the industrial camera simultaneously acquires the texture image of the oil cup at the current angle and the modulated structured light image. Subsequently, the operation execution unit drives the oil cup to rotate 180°, and the image acquisition unit acquires the image of the other side again.

[0052] S4: The image processing module comprehensively determines the laser engraving content, position, and engraving depth on the surface of the oil cup based on the texture image and three-dimensional morphology information, and outputs the determination result.

[0053] This step is an intelligent comprehensive judgment: This step is performed by the image processing module. Specifically, it includes: Content recognition: After preprocessing the texture image (such as contrast enhancement and binarization), the trained OCR algorithm and pattern template matching algorithm are used to recognize the laser-engraved characters, logos and other content to obtain the first sub-result (content correct / incorrect).

[0054] Depth Analysis: Based on the 3D point cloud data parsed from structured light images, the depth data of the laser-engraved stroke area is accurately extracted, and its average depth, depth variance, and other depth distribution characteristics are calculated. This feature is compared with a preset depth tolerance range (e.g., the depth value should be between 20-100μm, and the variance should be less than 10μm) to obtain a second sub-result (depth acceptable / unacceptable).

[0055] Fusion Decision: Perform a logical AND operation on the first and second sub-results mentioned above. That is, the image processing module 502 will only output a "qualified" final judgment result for the oil cup if both the "content is correct" and "depth is qualified" conditions are met simultaneously; if either condition is not met, an "unqualified" result and the specific defect type will be output.

[0056] S5: Based on the judgment result, the motion control module controls the operation execution unit and the material management unit to remove the unqualified oil cup and replace it with a qualified oil cup.

[0057] This step is a closed-loop material handling process, specifically: the motion control module controls the actuator based on the judgment result. If the oil cup is qualified, control the corresponding moving module to move it back to the original position of the fixture.

[0058] If an oil cup is defective, the system controls the moving module to transfer and release it to the defective product collection box. Simultaneously, the system records the coordinates of the empty space on the fixture and controls an idle moving module to pick up a qualified oil cup from the replenishment bin and precisely fill the empty space. This process achieves a closed-loop material flow of "instant inspection and rejection, instant replenishment of missing items."

[0059] S6: The process feedback module generates an adjustment instruction for the process parameters of the preceding laser engraving equipment based on the determination result.

[0060] This step involves process feedback and optimization, specifically: the process feedback module monitors and determines the results in real time. When the system detects that the laser engraving positions of N consecutive oil cups (e.g., 4) have shifted in the same direction exceeding a threshold, or that the engraving depth is consistently too shallow / too deep, a systematic deviation is identified. This module then automatically generates a control instruction package containing coordinate compensation or laser power adjustment values, which is sent to the CNC system of the preceding laser engraving equipment via a communication interface, guiding it to adaptively adjust process parameters and prevent quality problems at their source. Details are as follows: Adjustment condition determination: The system presets dynamic rules. For example, rule 1: "The 'positional offset' defect occurs more than 3 times in 10 consecutive products"; rule 2: "The quantitative parameter (depth difference) of the 'insufficient depth' defect shows a continuous increasing trend". When either rule is met, the process adjustment condition is determined to be met.

[0061] Command Generation and Transmission: Based on the triggered defect type, the system generates precise process parameter adjustment commands. For example, for "positional offset," the system calculates the coordinate offset compensation of the laser engraving machine's 200 galvanometer mirror based on the statistical average of the offset, generating a command of "X-axis compensation +0.05mm, Y-axis compensation -0.02mm." For "insufficient depth," commands such as "laser power increased by 5%" or "scanning speed reduced by 10%" are generated. These commands are transmitted to the host computer in real time via industrial Ethernet or RS-485 bus.

[0062] Effect Verification and Closed-Loop: After sending the adjustment command, the system will pay special attention to whether the relevant defects have disappeared or whether the quantitative parameters have returned to the normal range in the test results of several subsequent products. This data forms a closed loop, which is used to verify the adjustment effect and can further optimize the adjustment rule base, making the system more and more "intelligent" with use.

[0063] Example 2: System Expansion and Optimization The main difference between this embodiment and Example 1 is that, in some preferred embodiments, the system also has the following functions: Self-learning optimization function: The image processing module is connected to a sample database and a model update unit. For samples with low confidence in the judgment result or marked as "suspicious" (such as two conflicting sub-results), the system stores their image data, features, and the actual results after manual verification in the sample database. During non-production periods such as nighttime, the model update unit automatically starts, calling new samples from the database to incrementally train the recognition model and update the model parameters, enabling the system to continuously adapt to new laser engraving styles or defect patterns.

[0064] Data statistics and production insight functions: The control unit also includes a data statistics module. This module records information such as the location (jig row and column number), occurrence time, and production batch of all non-conforming products. Through analysis, various statistical reports can be generated, and when an abnormally high defect rate is found in a specific jig location or when a batch of materials experiences a concentrated outbreak of problems, it can automatically issue a production warning or instruct the replenishment module 402 to prioritize the use of inventory from other batches, thereby achieving dynamic optimization of production strategies.

[0065] Example 3: Method Example Based on the same inventive concept, this invention also provides an intelligent detection method for the laser engraving appearance of oil cups, which can be applied to any of the aforementioned system embodiments. The steps are fully encompassed in the detailed description of the system workflow (S1-S6) of Embodiment 1 above, and will not be repeated here.

[0066] The core of this method lies in the synergy of multimodal information fusion judgment, closed-loop material control and front-end process feedback, which realizes the comprehensiveness, efficiency and intelligence of the appearance quality inspection of oil cup laser engraving.

Claims

1. An intelligent inspection system for the appearance of laser-engraved oil cups, used for inspecting and sorting oil cups after laser engraving, characterized in that, The system includes: A conveying unit is used to carry and transport a fixture containing multiple oil cups; An operation execution unit is correspondingly set at the detection station of the conveying unit, used to pick up the oil cup from the conveying unit and move the oil cup to a detection position that is separated from the conveying unit; An image acquisition unit is set at the detection position. The image acquisition unit is a multimodal image acquisition component, used to simultaneously acquire the texture image and three-dimensional shape information of the oil cup located at the image acquisition station. The material management unit includes a collection module for receiving defective oil cups and a replenishment module for providing qualified oil cups; The control unit is communicatively connected to the conveying unit, image acquisition unit, operation execution unit, and material management unit, respectively. The control unit includes an image processing module, a motion control module, and a process feedback module. The image processing module is connected to the image acquisition unit and is used to acquire the texture image and three-dimensional morphology information, and to comprehensively judge the laser engraving quality on the surface of the oil cup based on preset judgment rules, and output the judgment result. The motion control module is connected to the image processing module, the operation execution unit, and the material management unit, and is used to control the operation execution unit and the material management unit to remove unqualified oil cups and replenish qualified oil cups according to the judgment result, so as to form a closed-loop material flow. The process feedback module is connected to the image processing module and is used to generate process parameter adjustment instructions for the preceding laser engraving equipment based on the laser engraving quality information in the judgment result.

2. The system according to claim 1, characterized in that, The multimodal image acquisition component includes at least one industrial camera and a structured light projector; the structured light projector is used to project a light pattern onto the surface of the oil cup, the industrial camera is used to acquire an image of the oil cup surface modulated by the light pattern, and the image processing module obtains the three-dimensional morphology information by analyzing the image acquired by the industrial camera.

3. The system according to claim 1, characterized in that, The operation execution unit includes a first moving module and a second moving module symmetrically arranged on both sides of the detection station of the conveying unit; each moving module includes a clamping mechanism and a rotary drive mechanism for driving the clamping mechanism to rotate.

4. The system according to claim 3, characterized in that, The rotary drive mechanism includes a rotary motor and a belt drive mechanism driven by the rotary motor; the clamping mechanism includes a plurality of grippers connected to the belt drive mechanism, and the rotary motor drives all grippers to rotate synchronously through the belt drive mechanism.

5. The system according to claim 1, characterized in that, The collection module is a defective product collection box, and the replenishment module is a replenishment bin; the motion control module controls the operation execution unit to put the defective oil cup into the defective product collection box, and to pick up the qualified oil cup from the replenishment bin to replenish the empty space of the conveying unit.

6. The system according to claim 1, characterized in that, The image processing module is connected to a sample database and a model update unit. The image processing module stores the sample data whose judgment results are marked as suspicious or require verification, along with their verification results, into the sample database. During non-production periods, the model update unit calls the data in the sample database to retrain the recognition model used for comprehensive judgment in the image processing module, thereby updating the model parameters.

7. The system according to claim 1, characterized in that, The image processing module's comprehensive assessment of the laser engraving quality on the oil cup surface includes the following steps: Analyze the texture image to identify the laser-engraved characters and patterns, and obtain the first sub-result; Analyze the three-dimensional topography information, extract the depth distribution characteristics of the laser-engraved area, and determine whether the engraving depth is qualified according to the preset depth standard to obtain the second sub-result; Based on the first sub-result and the second sub-result, perform a logical AND operation to output the final laser engraving quality judgment result; The final judgment result is qualified only if the content of the first sub-result is correctly identified and the depth judgment of the second sub-result is qualified.

8. The system according to claim 1, characterized in that, When the judgment results of multiple consecutive oil cups output by the image processing module all indicate that the laser engraving position has an offset in the same direction or the engraving depth has a deviation in the same trend, the process feedback module generates a control command including position compensation amount or laser power adjustment amount and sends it to the controller of the front-end laser engraving equipment.

9. The system according to claim 1, characterized in that, The control unit also includes a data statistics module, which is used to record the location, time or batch information of defective oil cups, and generate production early warning signals or send instructions to the motion control module to adjust the feeding strategy based on historical data statistical patterns.

10. A method for intelligent detection of the appearance of laser engraving on an oil cup, applied to the device as described in any one of claims 1-9, characterized in that, Includes the following steps: The conveying unit transports the fixture containing the oil cup to the testing station; The operation execution unit picks up the oil cup from the conveying unit and moves it to the detection position; The image acquisition unit simultaneously acquires the texture image and three-dimensional morphology information of the oil cup at the detection position; The image processing module comprehensively determines the laser engraving content, position, and engraving depth on the surface of the oil cup based on the texture image and three-dimensional morphology information, and outputs the determination result. Based on the determination result, the motion control module controls the operation execution unit and the material management unit to remove the unqualified oil cup and replace it with a qualified oil cup. The process feedback module generates an instruction to adjust the process parameters of the preceding laser engraving equipment based on the determination result.

Images