Machine vision-based connector terminal appearance defect detection equipment

Abstract

The invention relates to the technical field of connector terminal detection, in particular to connector terminal appearance defect detection equipment based on machine vision, which is characterized in that a plurality of connector terminals are poured into a feeding mechanism, and the feeding mechanism is used for sequentially arranging and forwards conveying the plurality of connector terminals; then comprehensive shooting detection is carried out on the connector terminals in the front direction, the rear direction, the upper direction and the lower direction through the appearance detection mechanism, while appearance detection is carried out, suspicious connector terminals and unqualified connector terminals are distinguished from the connector terminals through appearance detection processing, and conduction testing is carried out on the suspicious connector terminals through connection testing; and meanwhile, the unqualified connector terminal is directly moved out, so that the conduction test of products with qualified appearances and products with unqualified appearances is omitted, and the production process is further accelerated and the production efficiency is improved by fusing the appearance detection and the conduction test.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of connector terminal detection, more particularly, it relates to a connector terminal appearance defect detection equipment based on machine vision. BACKGROUND

[0002] In the electronic manufacturing industry, as the "bridge" for realizing signal and power transmission between circuits, the quality reliability of the connector terminal, the core component of the connector, directly determines the performance and service life of the entire connector and even the final product. The connector terminal needs to go through a series of complex processes such as precision stamping, bending, electroplating, etc. in production, and any slight deviation of the process parameters may introduce various types of defects.

[0003] At present, in the industry, the quality control at the single connector terminal stage widely adopts automatic optical detection equipment based on machine vision. This technology can efficiently detect "macro" and "surface" defects such as shape deformation, size out-of-tolerance, surface contamination, macro scratches, and stamping burrs by capturing two-dimensional images of the connector terminal surface through high-resolution cameras and using advanced image processing algorithms. This system has made remarkable achievements in improving production efficiency and reducing labor costs, and has become a standard configuration on modern production lines.

[0004] However, with the pursuit of extreme product reliability in the industry, especially in the "zero defect" fields of automotive electronics, aerospace, etc., the fundamental limitations of pure visual detection technology are increasingly evident, and it cannot truly judge the electrical performance of the connector. Therefore, the traditional approach is to wrap the connector terminal with a plastic shell, assemble it into a complete connector, and then perform final electrical verification. Test items usually include contact resistance, insulation resistance, voltage resistance, etc. The entire assembly of the connector is completed, but due to various forms of test items, the production progress is slow, which seriously affects the production efficiency. SUMMARY

[0005] In view of the shortcomings of the prior art, the present application aims to provide a connector terminal appearance defect detection equipment based on machine vision, which has the advantages of directly removing unqualified connector terminals, thereby saving the conduction test of appearance qualified products and appearance unqualified products, and utilizing the fusion of appearance detection and conduction test for detection, further accelerating the production process and improving the production efficiency.

[0006] The above technical purpose of the present application is achieved by the following technical scheme: a connector terminal appearance defect detection equipment based on machine vision, comprising a feeding mechanism capable of sequentially arranging and conveying a plurality of connector terminals; an appearance detection mechanism for comprehensive shooting and detection in multiple directions of the connector terminal, front, back, up and down; A continuity testing and inspection mechanism is used to test the continuity of connector terminals after they have been connected. It also includes testing methods for connector terminals: S1: The appearance inspection mechanism performs appearance defect inspection on the connector terminals conveyed by the feeding mechanism, and distinguishes the connector terminals from the suspicious connector terminals and the unqualified connector terminals in the appearance inspection process. S2: Use the connectivity test to perform a continuity test on the suspicious connector terminals, and at the same time remove the unqualified connector terminals directly.

[0007] Preferably, the feeding mechanism includes a feeding support platform, a feeding support frame disposed on the feeding support platform, a feeding cylinder installed on the feeding support frame, and a vibrating plate communicating with the outlet of the feeding cylinder. The connector terminals are received through the feeding cylinder, and the vibrating plate arranges the connector terminals in an orderly manner.

[0008] Preferably, the feeding support platform is provided with an output section, the conveying section is connected to the outlet of the vibratory feeder, and the output section is provided with a vibrator on the feeding support platform, which uses the amplitude of the vibrator to continuously output the connector terminals outward.

[0009] Preferably, the feeding support platform is rotatably connected to a rotating disk, a drive motor that drives the rotating disk to rotate, and suction claws that are evenly arranged along the circumferential edge of the rotating disk, and the suction claws grab the connector terminals flowing out along the output section.

[0010] Preferably, the appearance inspection mechanism includes multiple camera modules, which are arranged sequentially along the connector terminal conveying direction. The multiple camera modules face the top, bottom, left and right sides of the connector terminal respectively, to complete the inspection of the connector terminal in four directions: up, down, front and back.

[0011] Preferably, the continuity test mechanism includes a Kelvin probe that can abut against the connector terminal and a robotic arm for mounting the Kelvin probe. The robotic arm moves the Kelvin probe to perform a continuity test on the connector terminal.

[0012] Preferably, it also includes an output mechanism, through which the qualified connector terminals are output.

[0013] Preferably, the output mechanism includes a feeding support platform and a conveyor belt disposed on the feeding support platform. The conveyor belt moves intermittently, and the appearance inspection mechanism is distributed on the side of the conveyor belt.

[0014] Preferably, in step S1, the detection method uses an appearance inspection mechanism in conjunction with a multi-angle light source to take pictures of the key areas of the connector terminals. Based on the key areas of the connector terminals, the precise pixel coordinates and directions of suspected micro-scratches are identified, the boundaries of suspected micro-dimples are identified, and areas of texture abnormality caused by uneven stamping or electroplating are identified.

[0015] Preferably, based on the identification of suspicious connector terminals, a digital map of the suspicious area is output, and the output is verified by a connectivity testing and inspection agency based on the digital map of the suspicious area to identify defective connector terminals among the suspicious connector terminals.

[0016] In summary, the present invention has the following beneficial effects: multiple connector terminals are fed into the feeding mechanism, which then arranges and transports them forward in an orderly manner. A visual inspection mechanism then performs comprehensive imaging inspection of the connector terminals from multiple directions (front, back, top, and bottom). During this visual inspection, the connector terminals are distinguished as suspicious or unqualified. A continuity test is then performed on the suspicious connector terminals, while the unqualified connector terminals are directly removed. This saves time on conducting continuity tests on products that pass and fail visual inspection. Furthermore, the integration of visual inspection and continuity testing further accelerates the production process and improves production efficiency. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the device mechanism structure according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the testing method steps according to an embodiment of the present invention.

[0018] Reference numerals: 1. Feeding mechanism; 11. Feeding support platform; 12. Feeding support frame; 13. Feed cylinder; 14. Vibratory feeder; 15. Output section; 16. Vibratory feeder; 17. Rotary disc; 18. Suction claw; 2. Appearance inspection mechanism; 21. Camera module; 3. Connectivity test mechanism; 31. Robotic arm; 4. Output mechanism; 41. Unloading support platform; 42. Conveyor belt. Detailed Implementation

[0019] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0020] It should be noted that when a component is referred to as being "fixed to" or "set on" another component, it can be directly or indirectly attached to that other component. When a component is referred to as being "connected to" another component, it can be directly or indirectly connected to that other component.

[0021] It should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention.

[0022] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0023] A machine vision-based connector terminal appearance defect detection device, see [link to relevant documentation]. Figure 1 and Figure 2 It includes a feeding mechanism 1, which is used to transport multiple connector terminals forward in an orderly manner; Appearance inspection unit 2 performs comprehensive imaging inspection of the connector terminals from multiple directions, including front, back, top, and bottom. The continuity testing mechanism 3 is used to perform continuity testing on the connector terminals after they are connected. It also includes testing methods for connector terminals: S1: The appearance inspection mechanism 2 performs appearance defect inspection on the connector terminals conveyed by the feeding mechanism 1, and the appearance inspection process distinguishes the connector terminals from suspicious connector terminals and unqualified connector terminals. S2: Use the connectivity test to perform a continuity test on the suspicious connector terminals, and at the same time remove the unqualified connector terminals directly.

[0024] In this embodiment, multiple connector terminals are fed into the feeding mechanism 1, which then arranges and conveys them forward in an orderly manner. Afterwards, the appearance inspection mechanism 2 performs comprehensive imaging inspection of the connector terminals from multiple directions (front, back, top, and bottom). Simultaneously, the appearance inspection process distinguishes between suspicious and unqualified connector terminals. The continuity test is then used to perform a conductivity test on the suspicious connector terminals, while the unqualified connector terminals are directly removed. This saves time on conducting continuity tests on products with acceptable and unacceptable appearances. Furthermore, the integration of appearance inspection and continuity testing further accelerates the production process and improves production efficiency.

[0025] The feeding mechanism 1 includes a feeding support platform 11, a feeding support frame 12 on the feeding support platform 11, a feeding cylinder 13 installed on the feeding support frame 12, and a vibrating plate 1614 connected to the outlet of the feeding cylinder 13. The connector terminals are received by the feeding cylinder 13 and arranged in an orderly manner by the vibrating plate 1614.

[0026] Since each connector terminal needs to be tested independently, in order to improve testing efficiency, multiple disordered connector terminals are arranged in an orderly manner. The connector terminals are then poured into the feed cylinder 13, and the connector terminals are arranged in an orderly manner by the vibratory feeder 1614, thus providing the prerequisite for testing a single connector terminal.

[0027] An output section 15 is provided on the feeding support platform 11. The conveying section is connected to the outlet of the vibratory plate 1614. The output section 15 is provided with a vibrator on the feeding support platform 11, and the connector terminals are continuously output outward by means of the amplitude of the vibrator.

[0028] After the orderly arrangement is completed, the connector terminals are output through the output section 15, which uses the amplitude of the vibrator to continuously output the connector terminals outward.

[0029] Specifically, the feeding support platform 11 is rotatably connected to a rotating disk 17, a drive motor that drives the rotating disk 17 to rotate, and suction claws 18 that are evenly arranged along the circumferential edge of the rotating disk 17. The suction claws 18 grab the connector terminals flowing out along the output section 15.

[0030] The connector terminals output by the output unit 15 are received one by one by the rotating disk 17 and gripped by the suction claw 18. The suction claw 18 is a magnetic suction claw. While the rotating disk 17 is rotating, the appearance inspection mechanism 2 detects the downward direction of the connector terminals, and the continuity test mechanism 3 is used to perform a continuity test on the connector terminals after they are connected.

[0031] The appearance inspection mechanism 2 includes multiple camera modules 21, which are arranged sequentially along the connector terminal conveying direction. The multiple camera modules 21 face the top, bottom, left and right sides of the connector terminal respectively, to complete the inspection of the connector terminal in four directions: up, down and forward.

[0032] The continuity test mechanism 3 includes a Kelvin probe that can abut against the connector terminal and a robotic arm 31 for mounting the Kelvin probe. The robotic arm 31 is used to move the Kelvin probe to perform a continuity test on the connector terminal.

[0033] After a suspected suspicious connector terminal is detected in the appearance inspection below, it is verified by the continuity test inspection mechanism 3. If the Kelvin probe fails the continuity test of the connector terminal, the robot arm 31 will grab the unqualified connector terminal and remove it from the rotating disk 17.

[0034] The connector terminals that pass the test are output through the output mechanism.

[0035] The output mechanism includes a feeding support platform and a conveyor belt mounted on the feeding support platform. The conveyor belt moves intermittently, and the appearance inspection mechanism 2 is distributed on the side of the conveyor belt. The intermittent movement of the conveyor belt, combined with multiple camera modules 21, performs appearance inspections on the top, left, and right sides of the connector terminals.

[0036] During the inspection process, the inspection method in step S1 uses the appearance inspection mechanism 2 in conjunction with a multi-angle light source to take pictures of the key areas of the connector terminal. Based on the key areas of the connector terminal, the precise pixel coordinates and direction of suspected micro-scratches are identified, the boundaries of suspected micro-dimples are identified, and areas of texture abnormality caused by uneven stamping or electroplating are identified.

[0037] Furthermore, based on the identification of suspicious connector terminals, a digital map of the suspicious area is output. The connection testing and inspection mechanism 3 is used to verify the output based on the digital map of the suspicious area, and to identify the defective connector terminals among the suspicious connector terminals.

[0038] Specifically, the testing steps include Step 1: Automated and orderly feeding and positioning Centralized feeding: Connector terminals are poured into the feed cylinder 13 in batches.

[0039] Sorting and Orientation: The connector terminals enter the vibratory feeder 1614 by gravity. The vibratory feeder 1614 uses minute vibrations to arrange the disordered connector terminals in an orderly manner and orient them according to a preset direction, so that they enter the subsequent process with a consistent posture.

[0040] Linear transport: The arranged connector terminals enter the output section 15 from the outlet of the vibratory feeder 1614, typically along a linear vibratory track. The vibrator under the track ensures that the connector terminals can move forward smoothly and continuously.

[0041] Precision gripping and positioning: The drive motor rotates the rotary table 17 intermittently. The suction claw 18, mounted on the rotary table 17, precisely grips a connector terminal using magnetic attraction as it passes the output section 15. The rotary table 17 then sequentially transports the gripped connector terminals to the subsequent appearance inspection and electrical testing stations. This rotary design ensures high cycle time and production continuity.

[0042] Step Two: Multi-directional visual reconnaissance and preliminary assessment Collaborative Imaging: When the rotating disk 17 carrying the connector terminal stops at the visual inspection station, multiple camera modules 21 surrounding it take pictures of the top, bottom, left and right sides of the connector terminal simultaneously or sequentially under the cooperation of multi-angle light sources, such as low-angle light, coaxial light and backlight, to achieve coverage without blind spots.

[0043] Intelligent analysis and map generation: The image processing system analyzes the acquired images and quickly identifies macroscopic defects in the connector terminals, such as severe deformation, missing materials, and large-area damage, and marks them directly as NG.

[0044] Meanwhile, for more subtle features, the system performs fine recognition: locating the coordinates and direction of suspected micro-scratches, delineating the boundaries of suspected micro-dimples, and identifying areas of texture abnormality.

[0045] Ultimately, the system generates a "digital map of suspected areas" for each connector terminal. This map contains the type, precise location, and geometric features of all suspected defects. Connector terminals marked as suspicious are designated as "suspected connection connector terminals."

[0046] Step 3: Electrical Verification and Authoritative Diagnosis Precise positioning: For connector terminals visually marked as "suspicious," the rotary table 17 transports them to the electrical testing station. The robot arm 31, based on the received "suspicious area digital map," moves with a Kelvin probe to the precise coordinates specified on the map.

[0047] Continuity test verification: Kelvin probes use a four-wire measurement method to eliminate lead resistance and make constant force contact with suspicious areas of the connector terminals to perform micro-resistance measurement.

[0048] For scratches: the probe may perform multiple measurements along the scratch to check for abnormal spikes in the resistance value.

[0049] For pits: the probe will measure the resistance difference between the center of the pit and the surrounding normal area.

[0050] Final verdict: By comparing the measured resistance value with a preset functional threshold, it is verified whether the suspected defect truly affects the conductivity of the connector terminal. This leads to a final decision on the "suspected connector terminal": qualified or defective.

[0051] Step 4: Output of qualified products and rejection of defective products All connector terminals that pass the initial visual inspection and electrical testing are transported to the unloading position by the rotary table 17 and collected by the conveyor belt.

[0052] All defective connector terminals identified during initial visual inspection and those verified as substandard during electrical testing will be directly removed from the production line by the robot arm 31 at the corresponding workstation and placed into the defective product collection box.

[0053] Specifically, taking a connector terminal as the test object, the microscopic scratches and pits in its contact spring area are accurately detected, and it is determined whether these defects affect its electrical performance.

[0054] Camera Module 21: Camera 1: 5-megapixel monochrome camera with a telecentric lens, used for top 2D inspection. Camera 2: Same model, with a ring light source, used for bottom inspection. Light Source: High-brightness white strip light for side lighting to highlight scratches; coaxial light for dimensional measurement.

[0055] Kelvin probe instrument: Keithley DMM6500 6½-digit digital multimeter, supporting four-wire Kelvin measurement.

[0056] Probe: Dual probe power head, each driven by an independent servo motor. The probe is made of gold-plated tungsten steel with a tip curvature radius R=10μm.

[0057] Force sensor: integrated into the probe power head, with a range of 0-50gf and an accuracy of ±0.1gf.

[0058] It also includes a control and computing module, which uses an industrial PC (i7 CPU, 16GB RAM) and is equipped with custom control software developed with NI LabVIEW or Halcon and C#. This module is used for overall system control and performs image processing, motion control, data fusion, and logical judgment.

[0059] The specific steps are as follows: Step 1: System Initialization and Calibration Visual calibration: Using a standard checkerboard calibration board, perform camera intrinsic parameter distortion correction and extrinsic parameter pixel to world coordinate system conversion calibration.

[0060] Hand-eye calibration: Perform joint "vision-electrical measurement" calibration. Use a specially designed calibration block to visually identify the marker points on it, and at the same time use probes to physically contact these points to establish a precise mapping relationship between the visual coordinate system and the robot coordinate system.

[0061] Step 2: Visual reconnaissance and generation of "digital map of suspicious areas" Image acquisition: Rotary disk 17 transports the connector terminals to the vision station, and the top and bottom cameras take pictures under two different light sources, obtaining a total of 4 images.

[0062] Defect identification and feature extraction: Algorithm: A hybrid model of "traditional algorithm + deep learning" is adopted.

[0063] Traditional algorithms extract edges from high-contrast images and measure macroscopic dimensions such as the total length and pin width of connector terminals.

[0064] Deep learning: A trained U-Net semantic segmentation model is used to process the shrapnel region image and output a pixel-level classification map to identify "background", "normal area", "scratches" and "dents".

[0065] Generate a "digital map of suspicious areas" by converting the output of the deep learning model into a structured JSON file: { "terminal_id": "T20240527-001", "overall_visual_judgment": "SUSPECT", "defects": [ { "defect_id": "D1", "type": "SCRATCH", "confidence": 0.85, "contour_pixels": [[x1, y1], [x2, y2], ...], "bounding_box": {"x": 100, "y": 200, "w": 50, "h": 5}, "characteristics": { "length_um": 48.5, "direction_angle_deg": 12.5 } }, { "defect_id": "D2", "type": "PIT", "confidence": 0.78, "contour_pixels": [[x1, y1], [x2, y2], ...], "bounding_box": {"x": 150, "y": 180, "w": 10, "h": 10}, "characteristics": { "area_um2": 80.2 } } ] } Step 3: Platform movement: Rotary tray 17 transports the "suspicious" connector terminal to the electrical testing station.

[0066] Path planning: The control software analyzes the "digital map" to plan the probe movement path for the robotic arm 31. For example: first measure the scratch D1, then measure the dent D2.

[0067] Adaptive electrical measurement execution: For scratch D1: Control the dual probes to move to the starting point A and ending point B of the scratch, calculated based on bounding_box and direction_angle_deg.

[0068] The "dynamic path resistance scanning" method was used: probe A was fixed, and probe B started from point A and moved gradually along the scratch direction in 10μm steps, stopping and measuring the resistance once at each step. A set of resistance values ​​[R1, R2, ..., Rn] was obtained.

[0069] For pit D2: The "three-point reference comparison method" is adopted: probe A is moved to the center point C of the pit and the resistance R_center is measured. Then probes A and B are moved to the visually normal area around the pit respectively, and the resistances R_ref1, R_ref2, and R_ref3 at three different points are measured. The average reference resistance R_ref_avg is calculated.

[0070] Step 4: Data Fusion and Comprehensive Judgment (Core of Innovation) Rule 1: Scratch Judgment Calculate the maximum and average resistance R_max and resistance R_avg of the scratch path.

[0071] If (R_max / R_avg > 1.5) OR (standard deviation (R1..Rn) > 0.1 mΩ) THEN, the result is classified as "high-risk scratch", and the conclusion is NG.

[0072] Physical meaning: A sharp peak or violent fluctuation in resistance indicates the presence of micro-cracks or material inhomogeneity at the bottom of the scratch.

[0073] Rule 2: Determination of Dents Calculate the local resistance increment ΔR = R_center - R_ref_avg.

[0074] If IFΔR > 5 mΩ, THEN determines it to be a "functional pit", and the conclusion is NG.

[0075] Physical significance: The depth of the pit has damaged the conductive substrate or resulted in a significant reduction in the effective conductive cross-sectional area.

[0076] Step 5: Sorting and Data Archiving Based on the final judgment, the industrial control computer controls the air blowing valve to blow good and defective products into different material boxes when the connector terminals reach the unloading station.

[0077] All data, including images, maps, resistance curves, and judgment results, are stored in the database and bound to the connector terminal ID to achieve full lifecycle quality traceability.

[0078] The above embodiments are merely explanations of the present invention and are not intended to limit the present invention. After reading this specification, those skilled in the art can make modifications to these embodiments without contributing any inventive step, but as long as they are within the scope of the claims of the present invention, they are protected by patent law.

Claims

1. A machine vision-based connector terminal appearance defect detection device, characterized in that: It includes a feeding mechanism, which can be used to transport multiple connector terminals forward in an orderly manner; The appearance inspection agency conducts comprehensive photographic inspections of the connector terminals from multiple directions, including front, back, top, and bottom. A continuity testing and inspection mechanism is used to test the continuity of connector terminals after they have been connected. It also includes testing methods for connector terminals: S1: The appearance inspection mechanism performs appearance defect inspection on the connector terminals conveyed by the feeding mechanism, and distinguishes the connector terminals from the suspicious connector terminals and the unqualified connector terminals in the appearance inspection process. S2: Use the connectivity test to perform a continuity test on the suspicious connector terminals, and at the same time remove the unqualified connector terminals directly.

2. The connector terminal appearance defect detection device based on machine vision according to claim 1, characterized in that: The feeding mechanism includes a feeding support platform, a feeding support frame on the feeding support platform, a feeding cylinder installed on the feeding support frame, and a vibrating plate connected to the outlet of the feeding cylinder. The connector terminals are received through the feeding cylinder, and the vibrating plate arranges the connector terminals in an orderly manner.

3. The connector terminal appearance defect detection device based on machine vision according to claim 2, characterized in that: An output section is provided on the feeding support platform. The conveying section is connected to the outlet of the vibratory feeder. A vibrator is provided on the feeding support platform for the output section, and the connector terminals are continuously output outward by the amplitude of the vibrator.

4. The machine vision-based connector terminal appearance defect detection device according to claim 3, characterized in that: The feeding support platform is rotatably connected to a rotating disk, a drive motor that drives the rotating disk to rotate, and suction claws that are evenly arranged along the circumferential edge of the rotating disk. The suction claws grab the connector terminals flowing out along the output section.

5. The connector terminal appearance defect detection device based on machine vision according to claim 1, characterized in that: The appearance inspection mechanism includes multiple camera modules, which are arranged sequentially along the connector terminal conveying direction. The multiple camera modules face the top, bottom, left and right sides of the connector terminal respectively, to complete the inspection of the connector terminal in four directions: up, down, front and back.

6. The machine vision-based connector terminal appearance defect detection device according to claim 1, characterized in that: The connectivity testing mechanism includes a Kelvin probe that can abut against the connector terminals and a robotic arm for mounting the Kelvin probe. The robotic arm moves the Kelvin probe to perform a continuity test on the connector terminals.

7. The machine vision-based connector terminal appearance defect detection device according to claim 1, characterized in that: Also includes The output mechanism outputs the qualified connector terminals.

8. The connector terminal appearance defect detection device based on machine vision according to claim 7, characterized in that: The output mechanism includes a feeding support platform and a conveyor belt disposed on the feeding support platform. The conveyor belt moves intermittently, and the appearance inspection mechanism is distributed on the side of the conveyor belt.

9. A machine vision-based connector terminal appearance defect detection device according to claim 1, characterized in that: In step S1, the detection method uses an appearance inspection mechanism in conjunction with a multi-angle light source to take pictures of the key areas of the connector terminals. Based on the key areas of the connector terminals, the precise pixel coordinates and directions of suspected micro-scratches are identified, the boundaries of suspected micro-dimples are identified, and areas of texture abnormality caused by uneven stamping or electroplating are identified.

10. A machine vision-based connector terminal appearance defect detection device according to claim 9, characterized in that: Based on the identification of suspicious connector terminals, a digital map of the suspicious area is output. The suspicious area digital map is then used to verify the output using a connectivity testing and inspection agency to identify defective connector terminals among the suspicious connector terminals.

Images