Cable electrical performance detection equipment and method based on visual identification

Abstract

The invention discloses a cable electrical performance detection device and method based on visual identification. The device comprises a mechanical arm module, a machine visual identification module and an industrial personal computer. The machine vision recognition module is installed on the mechanical arm module and used for collecting an image of the electric connector, matching and comparing the collected image with a pre-established vision model library, recognizing the model of the electric connector and obtaining space coordinates and normal information of each contact pin or jack; the probe end of the probe test module is in direct contact with the tested cable electric connector and forms a test loop with a circuit used for electrical performance test, a prefabricated tool cable is not arranged in the middle to serve as a middle connecting piece, electrical performance detection of various types of cables can be completed under the same equipment structure, and therefore the tool manufacturing and management cost is remarkably reduced, and the test efficiency is improved. And the detection efficiency, the consistency and the data traceability are improved.

Description

Technical Field

[0001] This disclosure relates to the field of cable testing technology, and in particular to a cable electrical performance testing device and method based on visual recognition. Background Technology

[0002] After cable manufacturing is completed, it typically requires testing for electrical properties such as continuity, insulation resistance, and withstand voltage to ensure product quality and safe use. Current cable production is generally divided into two categories: batch production projects and research and development projects. For batch production projects, a "cable testing equipment + tooling cable" approach is generally used. This involves pre-fabricated specialized tooling cables working in conjunction with a general-purpose testing bench to achieve efficient electrical performance testing of a large batch of cables. For research and development projects, due to the large variety of cable types, small quantities, and frequent structural changes, if the method of manufacturing tooling cables type by type is adopted, it requires material procurement, tooling design and manufacturing, and acceptance, resulting in high costs, long cycles, and a wide variety of tooling types, significantly increasing the difficulty of management and maintenance.

[0003] Therefore, during the research and development phase, handheld instruments such as multimeters, insulation tests, and withstand voltage testers are typically used to manually test each point on the cable connector according to the process documents. This method has the following drawbacks:

[0004] 1. Low efficiency: The testing time is long, making it difficult to meet the needs of frequent modifications and rapid verification during the development phase;

[0005] 2. Reliance on operator experience: Connector points are numerous and complexly distributed, and manual point-by-point contact is prone to omissions, errors, or poor contact.

[0006] 3. Poor consistency and traceability: Manual operation makes it difficult to maintain complete consistency in each testing process, and test results are not easy to trace, which is not conducive to quality problem analysis and responsibility division;

[0007] 4. High investment in tooling: To improve efficiency, a large number of special tooling cables still need to be manufactured, which is not commensurate with the investment and benefits for small-batch research and development projects.

[0008] In summary, the current technology lacks a general-purpose testing device that does not require the separate fabrication of tooling cables for each type of cable, can automatically identify cable connector locations through machine vision, drive a robotic arm to automatically complete probe contact and electrical performance testing, and has good generalization ability and data traceability function.

[0009] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0010] The purpose of this disclosure is to provide a cable electrical performance testing device and method based on visual recognition, thereby overcoming, at least to some extent, one or more problems caused by the limitations and defects of related technologies.

[0011] To achieve the above objectives, the present invention provides a cable electrical performance testing device based on visual recognition, comprising:

[0012] The robotic arm module is used to install machine vision recognition modules and probe testing modules, and moves in space relative to the electrical connector of the cable under test under control commands.

[0013] A machine vision recognition module is installed on the robotic arm module to acquire images of the electrical connector and match and compare the acquired images with a pre-established visual model library to identify the model of the electrical connector and obtain the spatial coordinates and normal information of each pin or socket.

[0014] The probe testing module is installed at the end of the robotic arm module. It is used to move to the spatial coordinates under the drive of the robotic arm module and contact the corresponding pin or socket. When it makes electrical contact with the test point, it applies a test signal to the test point and collects the response signal to perform at least one electrical performance test of the circuit, insulation resistance, and withstand voltage and output the test result.

[0015] An industrial control computer is communicatively connected to the robotic arm module, the machine vision recognition module, and the probe testing module. It is used to run the test system software to control vision recognition, robotic arm movement, electrical performance testing, and test data recording and display.

[0016] In this design, the probe end of the probe test module directly contacts the electrical connector of the cable under test and forms a test loop with the circuit used for electrical performance testing. No prefabricated tooling cable is set in the middle as an intermediate connector, so that the electrical performance testing of different types of cables can be completed under the same equipment structure.

[0017] Furthermore, the machine vision recognition module includes an industrial camera, a light source, and an image processing unit, wherein the image processing unit is configured as follows:

[0018] After completing the camera intrinsic parameter calibration and the camera-robotic arm and / or workstation coordinate system hand-eye calibration, the industrial camera is controlled to acquire target images according to the predetermined exposure, gain, polarization and light field parameters.

[0019] The acquired target image is subjected to denoising, distortion correction, brightness normalization and specular / reflection suppression to obtain a standardized image frame for recognition.

[0020] Furthermore, the image processing unit is also configured to:

[0021] Based on the trained object detection network, inference is performed on the standardized image to output candidate regions and instance masks containing the connector shape;

[0022] Instance segmentation or contour extraction based on edge detection and morphological operations is performed within the candidate region to obtain the outer contour and inner cavity contour of the connector, and at least one key element among the positioning key, snap-fit ​​or guide surface is identified.

[0023] Perform parameterized fitting on the repeating hole / pin array and number the pins or holes in a predetermined order.

[0024] Furthermore, the visual model library is constructed through the following steps:

[0025] Image acquisition and camera calibration were performed on multiple connector samples to obtain standardized image frames and acquired metadata.

[0026] Candidate region detection and contour extraction methods are used to obtain the outer contour and inner cavity contour of each connector, and to identify key geometric elements such as positioning keys, guide surfaces and snap fasteners.

[0027] Number each pin or socket, determine the nominal three-dimensional coordinates and normal information of each point in the model coordinate system, and establish a mapping with the manufacturer's naming table;

[0028] The original image, candidate region, contour data, number table, array parameters, coordinate system definition, and calibration and acquisition metadata are structured, packaged, and stored in the database. An index is created according to connector family / type / variant to form visual model entries.

[0029] Furthermore, after acquiring the two-dimensional observation information of key points, the machine vision recognition module is further configured as follows:

[0030] Call the visual model entry corresponding to the identified connector family / type / variant to obtain the key geometric points of the connector and the three-dimensional nominal coordinates of each pin / hole in the model coordinate system;

[0031] The pose of the connector relative to the camera is determined based on the camera intrinsic parameters and the correspondence between key points.

[0032] Complete the stepwise coordinate transformation from the model coordinate system, camera coordinate system, workstation coordinate system and / or robot arm base coordinate system to obtain the spatial coordinates and normal information of each pin or hole in the robot arm base coordinate system, and use the spatial coordinates and normal information as input for robot arm path planning.

[0033] Furthermore, the device also includes a clamping module, the clamping module comprising:

[0034] Adjustable guide rails that can be arranged along the length direction;

[0035] The movable clamping block is provided on the guide rail, and the position of the movable clamping block can be adjusted along the guide rail to adapt to the shape of electrical connectors of different widths or lengths.

[0036] The elastic pad layer disposed on the contact surface between the movable clamping block and the electrical connector is used to improve clamping stability and reduce local stress during the clamping process.

[0037] Furthermore, the clamping module is configured to form a flexible universal tooling so that various types of electrical connectors can be clamped with simple adjustments.

[0038] Furthermore, the probe testing module includes a probe module and an electrical performance testing module;

[0039] The probe module includes at least one probe and a probe mounting base. The probe mounting base is rigidly connected to the end of the robotic arm, and the probe extends out of the probe mounting base in a preset direction.

[0040] The electrical performance testing module is electrically connected to the probe module and is used to apply test signals to the probe and acquire response signals.

[0041] Furthermore, when the industrial control computer controls the robotic arm module to drive the probe testing module to move to the target spatial coordinates, it is configured to set the angle between the probe contact direction and the normal information within a preset tolerance range, so as to reduce contact deviation and improve the reliability of electrical contact with the pin or socket.

[0042] Furthermore, the device also includes a safety protection device, which includes a safety light curtain and an emergency stop button. When the safety light curtain detects personnel entering the detection area and / or the emergency stop button is triggered, the industrial control computer sends a stop command to the robotic arm module and the probe testing module through the device control module of the test system software to stop the movement of the robotic arm and shut down the high voltage output of the probe testing module. The current detection status is recorded by the data storage and traceability module.

[0043] Another aspect of the present invention provides a method for testing the electrical properties of cables using the above-described equipment, comprising:

[0044] 1) Workpiece fixing steps: Fix the electrical connector of the cable to be tested to the clamping module and / or the fixed station;

[0045] 2) Image acquisition and preprocessing steps: The machine vision recognition module is moved to the preset shooting position by the robotic arm module, and the target image of the electrical connector is acquired by the industrial camera. The target image is then processed by noise reduction, distortion correction, brightness normalization and highlight / reflection suppression to obtain a standardized image frame.

[0046] 3) Image recognition and point analysis steps: Based on the target detection network and instance segmentation / contour extraction, candidate regions containing electrical connectors are detected from the standardized image frames, positioning keys, buckles, guide surfaces and repeating hole / pin arrays are identified, and the visual model library is called to perform matching comparison to determine the model of the electrical connector under test and the three-dimensional spatial coordinates and normal information of each pin or socket.

[0047] 4) Path planning and probe positioning steps: The industrial control computer plans the movement trajectory of the robotic arm and the contact posture of the probe based on the three-dimensional spatial coordinates and normal information, and controls the robotic arm module to drive the probe testing module to move to each test point in sequence and make contact with it.

[0048] 5) Electrical performance testing and result judgment steps: After the probe test module makes reliable contact with each test point, the probe test module performs at least one electrical performance test among the following: continuity, insulation resistance, and withstand voltage. The test results are returned to the industrial control computer, which judges the test results according to the preset pass criteria and completes the test data recording and display.

[0049] Furthermore, the image recognition and point location analysis steps also include:

[0050] Multi-scale appearance features and geometric key point responses are extracted from candidate regions. Similarity calculation is performed by metric retrieval and visual prototypes and key geometric templates of corresponding family / type / variant entries in the visual model library to complete connector model determination.

[0051] After obtaining the 3D definition provided by the 2D observation and vision model library of key geometric points, the pose of the connector relative to the camera is solved, and the stepwise coordinate transformation from the model coordinate system, camera coordinate system, workstation and / or robot arm base coordinate system is completed to obtain the target space coordinates and normal information of each pin or socket in the robot arm base coordinate system.

[0052] This invention integrates a machine vision recognition module and a probe testing module at the end of a robotic arm, along with a vision model library. An industrial control computer coordinates vision recognition, motion control, and electrical performance testing, enabling the equipment to automatically identify the three-dimensional spatial coordinates and normals of different cable connector models and their pins / holes. The robotic arm then automatically and accurately positions the probes to the target location and completes electrical performance tests such as continuity, insulation resistance, and withstand voltage. The entire test circuit is directly formed by the probe testing module and the connector under test, eliminating the need for prefabricated tooling cables. This allows for universal testing of multiple cable models within the same equipment structure, significantly reducing tooling design and management costs, improving point positioning accuracy and testing efficiency, reducing human error, and making the cable electrical performance testing process highly automated, programmable, and easy to integrate.

[0053] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0054] The above and other features and advantages of this disclosure will become more apparent from the detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0055] Figure 1 A system block diagram of a vision-based cable electrical performance testing device according to an exemplary embodiment of the present disclosure is shown;

[0056] Figure 2 A schematic diagram of the external structure of a cable electrical performance testing device according to an exemplary embodiment of the present disclosure is shown.

[0057] Figure 3 A schematic flowchart of a cable electrical performance testing method according to an exemplary embodiment of the present disclosure is shown;

[0058] Figure 4 A schematic diagram illustrating the effect of visual recognition and contour extraction of cable connector locations according to an exemplary embodiment of the present disclosure is shown. Detailed Implementation

[0059] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted.

[0060] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details described, or other methods, components, materials, apparatuses, steps, etc., can be employed. In other instances, well-known structures, methods, apparatuses, implementations, materials, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0061] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, or in one or more software-hardened modules, or in different network and / or processor devices and / or microcontroller devices.

[0062] In this example embodiment, as Figure 1 and 2 As shown, one aspect of the present invention provides a cable electrical performance testing device based on vision recognition, including a robotic arm module 1, a machine vision recognition module 2, a probe testing module, and an industrial control computer 3.

[0063] Robotic arm module 1 is used to install machine vision recognition module 2 and probe testing module, and moves in space relative to the electrical connector of the cable under test under control commands;

[0064] The machine vision recognition module 2 is installed on the robotic arm module 1. It is used to collect images of the electrical connector and match and compare the collected images with a pre-established visual model library to identify the model of the electrical connector and obtain the spatial coordinates and normal information of each pin or socket.

[0065] The probe testing module is installed at the end of the robotic arm module 1. It is used to move to the spatial coordinates under the drive of the robotic arm module 1 and contact the corresponding pin or socket. When it makes electrical contact with the test point, it applies a test signal to the test point and collects the response signal to perform at least one electrical performance test of the circuit, insulation resistance, and withstand voltage and output the test result.

[0066] The industrial control computer 3 is communicatively connected to the robotic arm module 1, the machine vision recognition module 2, and the probe testing module, and is used to run the test system software to control vision recognition, robotic arm movement, electrical performance testing, and test data recording and display.

[0067] In this design, the probe end of the probe test module directly contacts the electrical connector of the cable under test and forms a test loop with the circuit used for electrical performance testing. No prefabricated tooling cable is set in the middle as an intermediate connector, so that the electrical performance testing of different types of cables can be completed under the same equipment structure.

[0068] In one embodiment of the present invention, the device further includes a clamping module 4, the clamping module 4 comprising: an adjustable guide rail that can be arranged along the length direction; a movable clamping block disposed on the guide rail, the position of the movable clamping block being adjustable along the guide rail to adapt to the shape of electrical connectors of different widths or lengths; an elastic pad layer disposed on the contact surface between the movable clamping block and the electrical connector, for improving clamping stability and reducing local stress during clamping; and the clamping module 4 is configured to form a flexible universal tooling so that clamping of various types of electrical connectors can be completed with simple adjustments.

[0069] In one embodiment of the present invention, the probe testing module includes a probe module 5 and an electrical performance testing module 6;

[0070] The probe module 5 includes at least one probe and a probe mounting base. The probe mounting base is rigidly connected to the end of the robotic arm, and the probe extends out of the probe mounting base in a preset direction.

[0071] The electrical performance testing module 6 is electrically connected to the probe module 5 and is used to apply a test signal to the probe and acquire a response signal.

[0072] Furthermore, when the industrial control computer 3 controls the robotic arm module 1 to drive the probe testing module to move to the target spatial coordinates, it is configured to set the angle between the probe contact direction and the normal information within a preset tolerance range, so as to reduce contact deviation and improve the reliability of electrical contact with the pin or socket.

[0073] In one embodiment of the present invention, the device further includes a safety protection device 7, which includes a safety light curtain and an emergency stop button. When the safety light curtain detects that a person has entered the detection area and / or the emergency stop button is triggered, the industrial control computer 3 sends a stop command to the robotic arm module 1 and the probe test module through the device control module of the test system software to stop the movement of the robotic arm and shut down the high voltage output of the probe test module. The current detection status is recorded by the data storage and traceability module.

[0074] In one embodiment of the present invention, the device further includes a test bench 8, which is a rigid support platform used to install the above-mentioned modules and provide a unified mechanical reference surface.

[0075] In one embodiment of the present invention, the machine vision recognition module 2 includes an industrial camera, a light source, and an image processing unit, wherein the image processing unit is configured as follows:

[0076] After completing the camera intrinsic parameter calibration and the camera-robotic arm and / or workstation coordinate system hand-eye calibration, the industrial camera is controlled to acquire target images according to the predetermined exposure, gain, polarization and light field parameters.

[0077] The acquired target image is subjected to denoising, distortion correction, brightness normalization and specular / reflection suppression to obtain a standardized image frame for recognition.

[0078] In one embodiment of the present invention, the image processing unit is further configured to:

[0079] Based on the trained object detection network, inference is performed on the standardized image to output candidate regions and instance masks containing the connector shape;

[0080] Instance segmentation or contour extraction based on edge detection and morphological operations is performed within the candidate region to obtain the outer contour and inner cavity contour of the connector, and at least one key element among the positioning key, snap-fit ​​or guide surface is identified.

[0081] Perform parameterized fitting on the repeating hole / pin array and number the pins or holes in a predetermined order.

[0082] In one embodiment of the present invention, the visual model library is constructed through the following steps:

[0083] Image acquisition and camera calibration were performed on multiple connector samples to obtain standardized image frames and acquired metadata.

[0084] Candidate region detection and contour extraction methods are used to obtain the outer contour and inner cavity contour of each connector, and to identify key geometric elements such as positioning keys, guide surfaces and snap fasteners.

[0085] Number each pin or socket, determine the nominal three-dimensional coordinates and normal information of each point in the model coordinate system, and establish a mapping with the manufacturer's naming table;

[0086] The original image, candidate region, contour data, number table, array parameters, coordinate system definition, and calibration and acquisition metadata are structured, packaged, and stored in the database. An index is created according to connector family / type / variant to form visual model entries.

[0087] In one embodiment of the present invention, after acquiring the two-dimensional observation information of key points, the machine vision recognition module 2 is further configured as follows:

[0088] Call the visual model entry corresponding to the identified connector family / type / variant to obtain the key geometric points of the connector and the three-dimensional nominal coordinates of each pin / hole in the model coordinate system;

[0089] The pose of the connector relative to the camera is determined based on the camera intrinsic parameters and the correspondence between key points.

[0090] Complete the stepwise coordinate transformation from the model coordinate system, camera coordinate system, workstation coordinate system and / or robot arm base coordinate system to obtain the spatial coordinates and normal information of each pin or hole in the robot arm base coordinate system, and use the spatial coordinates and normal information as input for robot arm path planning.

[0091] like Figure 3 As shown, the method for testing the electrical performance of cables using the above-mentioned equipment according to the present invention includes:

[0092] Step S310: Workpiece fixing step: Fix the electrical connector of the cable to be tested to the clamping module and / or the fixed station;

[0093] Step S320 Image Acquisition and Preprocessing: The machine vision recognition module is moved to the preset shooting position by the robotic arm module, and the target image of the electrical connector is acquired by the industrial camera. The target image is then processed for noise reduction, distortion correction, brightness normalization and highlight / reflection suppression to obtain a standardized image frame.

[0094] Step S330 Image recognition and point analysis step: Based on the target detection network and instance segmentation / contour extraction, the candidate region containing the electrical connector is detected from the standardized image frame, the positioning key, buckle, guide surface and repeating hole / pin array are identified, and the visual model library is called to perform matching comparison to determine the model of the electrical connector under test and the three-dimensional spatial coordinates and normal information of each pin or socket.

[0095] Step S340 Path planning and probe positioning: The industrial control computer plans the movement trajectory of the robotic arm and the contact posture of the probe based on the three-dimensional spatial coordinates and normal information, and controls the robotic arm module to drive the probe testing module to move to each test point in sequence and make contact with it.

[0096] Step S350 Electrical performance testing and result determination: After the probe test module makes reliable contact with each test point, the probe test module performs at least one electrical performance test among the following: continuity, insulation resistance, and withstand voltage. The test results are returned to the industrial control computer, which determines the test results according to the preset pass / fail criteria and completes the recording and display of test data.

[0097] In one embodiment of the present invention, step S330 further includes:

[0098] Multi-scale appearance features and geometric key point responses are extracted from candidate regions. Similarity calculation is performed by metric retrieval and visual prototypes and key geometric templates of corresponding family / type / variant entries in the visual model library to complete connector model determination.

[0099] After obtaining the 3D definition provided by the 2D observation and vision model library of key geometric points, the pose of the connector relative to the camera is solved, and the stepwise coordinate transformation from the model coordinate system, camera coordinate system, workstation and / or robot arm base coordinate system is completed to obtain the target space coordinates and normal information of each pin or socket in the robot arm base coordinate system.

[0100] Figure 4 This is a schematic diagram illustrating the visual recognition and contour extraction effect of the cable connector location in this invention. An industrial camera acquires the original image of the cable connector on the clamping module and performs preprocessing such as noise reduction, distortion correction, and brightness normalization. Figure 4Candidate regions containing connectors are obtained; within these candidate regions, the outer and inner contours of the connectors are identified using object detection, instance segmentation, and / or contour extraction algorithms, and the positions of each pin or socket are displayed as marked points in the figure.

[0101] In summary, compared with the prior art, the present invention has at least the following beneficial effects:

[0102] 1. Eliminate dedicated tooling cables

[0103] The probe end of the probe test module directly contacts the connector under test. The internal circuit of the probe test module and the cable under test form a test loop. There is no need for prefabricated tooling cables in between, which greatly reduces the cost of tooling design, processing and inventory, and simplifies tooling management.

[0104] 2. Strong adaptability to multiple varieties and small batches

[0105] The adjustable clamping module forms a flexible and universal tooling, which, together with the expandable vision model library and recognition algorithm, can quickly adapt to connectors of different structures and specifications. No hardware modification is required; new cable models can be supported simply by expanding the model entries on the software side.

[0106] 3. Automatic 3D point analysis, high positioning accuracy

[0107] This invention utilizes visual algorithms such as target detection, instance segmentation or contour extraction, multi-scale feature and geometric key point matching, combined with a visual model library containing nominal three-dimensional coordinates, to automatically obtain the spatial coordinates and normal information of each pin / hole through pose solving and coordinate transformation, which significantly improves the positioning accuracy and stability compared to traditional manual surveying or manual teaching methods.

[0108] 4. The probe attitude is constrained by the normal direction, resulting in good contact reliability.

[0109] During path planning, limiting the angle between the probe contact direction and the point normal within a preset tolerance range can effectively reduce the risk of poor contact or damage to the pin caused by probe misalignment, thereby improving the reliability of electrical connections and the repeatability of test results.

[0110] 5. The testing process is automated and highly efficient.

[0111] By using an industrial control computer to uniformly schedule machine vision recognition, robotic arm movement, and probe testing modules, an automated process is achieved from clamping, recognition, probe alignment to testing and judgment. Compared with manual point-by-point inspection, this significantly shortens the inspection time for each cable.

[0112] 6. Strong data traceability

[0113] The testing system software records the identification results, detection configuration, process parameters, and test results throughout the entire process. It can query and analyze historical data, support the tracking and location of quality problems, and meet the needs of process control and quality auditing.

[0114] 7. High security

[0115] By linking the safety light curtain and emergency stop button with the industrial control computer, the safety status of personnel is monitored in real time during the movement of the robotic arm and the high-voltage test. If any abnormality occurs, the movement and high-voltage output will be stopped immediately, effectively reducing the risk of equipment operation.

[0116] It should be noted that although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0117] Furthermore, the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of the present invention, and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0118] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0119] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A cable electrical performance testing device based on visual recognition, characterized in that, include: The robotic arm module is used to install machine vision recognition modules and probe testing modules, and moves in space relative to the electrical connector of the cable under test under control commands. A machine vision recognition module is installed on the robotic arm module to acquire images of the electrical connector and match and compare the acquired images with a pre-established visual model library to identify the model of the electrical connector and obtain the spatial coordinates and normal information of each pin or socket. The probe testing module is installed at the end of the robotic arm module. It is used to move to the spatial coordinates under the drive of the robotic arm module and contact the corresponding pin or socket. When it makes electrical contact with the test point, it applies a test signal to the test point and collects the response signal to perform at least one electrical performance test of the circuit, insulation resistance, and withstand voltage and output the test result. An industrial control computer is communicatively connected to the robotic arm module, the machine vision recognition module, and the probe testing module. It is used to run the test system software to control vision recognition, robotic arm movement, electrical performance testing, and test data recording and display. In this design, the probe end of the probe test module directly contacts the electrical connector of the cable under test and forms a test loop with the circuit used for electrical performance testing. No prefabricated tooling cable is set in the middle as an intermediate connector, so that the electrical performance testing of different types of cables can be completed under the same equipment structure.

2. The device according to claim 1, characterized in that, The machine vision recognition module includes an industrial camera, a light source, and an image processing unit, wherein the image processing unit is configured as follows: After completing the camera intrinsic parameter calibration and the camera-robotic arm and / or workstation coordinate system hand-eye calibration, the industrial camera is controlled to acquire target images according to the predetermined exposure, gain, polarization and light field parameters. The acquired target image is subjected to denoising, distortion correction, brightness normalization and specular / reflection suppression to obtain a standardized image frame for recognition.

3. The device according to claim 1, characterized in that, The image processing unit is further configured to: Based on the trained object detection network, inference is performed on the standardized image to output candidate regions and instance masks containing the connector shape; Instance segmentation or contour extraction based on edge detection and morphological operations is performed within the candidate region to obtain the outer contour and inner cavity contour of the connector, and at least one key element among the positioning key, snap-fit ​​or guide surface is identified. Perform parameterized fitting on the repeating hole / pin array and number the pins or holes in a predetermined order.

4. The device according to claim 1, characterized in that, The visual model library is constructed through the following steps: Image acquisition and camera calibration were performed on multiple connector samples to obtain standardized image frames and acquired metadata. Candidate region detection and contour extraction methods are used to obtain the outer contour and inner cavity contour of each connector, and to identify key geometric elements such as positioning keys, guide surfaces and snap fasteners. Number each pin or socket, determine the nominal three-dimensional coordinates and normal information of each point in the model coordinate system, and establish a mapping with the manufacturer's naming table; The original image, candidate region, contour data, number table, array parameters, coordinate system definition, and calibration and acquisition metadata are structured, packaged, and stored in the database. An index is created according to connector family / type / variant to form visual model entries.

5. The device according to claim 1, characterized in that, After acquiring the two-dimensional observation information of key points, the machine vision recognition module is further configured as follows: Call the visual model entry corresponding to the identified connector family / type / variant to obtain the key geometric points of the connector and the three-dimensional nominal coordinates of each pin / hole in the model coordinate system; The pose of the connector relative to the camera is determined based on the camera intrinsic parameters and the correspondence between key points. Complete the stepwise coordinate transformation from the model coordinate system, camera coordinate system, workstation coordinate system and / or robot arm base coordinate system to obtain the spatial coordinates and normal information of each pin or hole in the robot arm base coordinate system, and use the spatial coordinates and normal information as input for robot arm path planning.

6. The device according to claim 1, characterized in that, The device further includes a clamping module, the clamping module comprising: Adjustable guide rails that can be arranged along the length direction; The movable clamping block is provided on the guide rail, and the position of the movable clamping block can be adjusted along the guide rail to adapt to the shape of electrical connectors of different widths or lengths. The elastic pad layer disposed on the contact surface between the movable clamping block and the electrical connector is used to improve clamping stability and reduce local stress during the clamping process. Furthermore, the clamping module is configured to form a flexible universal tooling so that various types of electrical connectors can be clamped with simple adjustments.

7. The device according to claim 1, characterized in that, The probe testing module includes a probe module and an electrical performance testing module; The probe module includes at least one probe and a probe mounting base. The probe mounting base is rigidly connected to the end of the robotic arm, and the probe extends out of the probe mounting base in a preset direction. The electrical performance testing module is electrically connected to the probe module and is used to apply test signals to the probe and acquire response signals. Furthermore, when the industrial control computer controls the robotic arm module to drive the probe testing module to move to the target spatial coordinates, it is configured to set the angle between the probe contact direction and the normal information within a preset tolerance range, so as to reduce contact deviation and improve the reliability of electrical contact with the pin or socket.

8. The device according to claim 1, characterized in that, The equipment also includes a safety protection device, which includes a safety light curtain and an emergency stop button. When the safety light curtain detects personnel entering the detection area and / or the emergency stop button is triggered, the industrial control computer sends a stop command to the robotic arm module and the probe testing module through the equipment control module of the test system software to stop the movement of the robotic arm and shut down the high voltage output of the probe testing module. The current detection status is recorded by the data storage and traceability module.

9. A method for testing the electrical properties of cables using the equipment described in any one of claims 1 to 8, characterized in that, include: 1) Workpiece fixing steps: Fix the electrical connector of the cable to be tested to the clamping module and / or the fixed station; 2) Image acquisition and preprocessing steps: The machine vision recognition module is moved to the preset shooting position by the robotic arm module, and the target image of the electrical connector is acquired by the industrial camera. The target image is then processed by noise reduction, distortion correction, brightness normalization and highlight / reflection suppression to obtain a standardized image frame. 3) Image recognition and point analysis steps: Based on the target detection network and instance segmentation / contour extraction, candidate regions containing electrical connectors are detected from the standardized image frames, positioning keys, buckles, guide surfaces and repeating hole / pin arrays are identified, and the visual model library is called to perform matching comparison to determine the model of the electrical connector under test and the three-dimensional spatial coordinates and normal information of each pin or socket. 4) Path planning and probe positioning steps: The industrial control computer plans the movement trajectory of the robotic arm and the contact posture of the probe based on the three-dimensional spatial coordinates and normal information, and controls the robotic arm module to drive the probe testing module to move to each test point in sequence and make contact with it. 5) Electrical performance testing and result judgment steps: After the probe test module makes reliable contact with each test point, the probe test module performs at least one electrical performance test among the following: continuity, insulation resistance, and withstand voltage. The test results are returned to the industrial control computer, which judges the test results according to the preset pass criteria and completes the test data recording and display.

10. The method according to claim 9, characterized in that, The image recognition and point location analysis steps also include: Multi-scale appearance features and geometric key point responses are extracted from candidate regions. Similarity calculation is performed by metric retrieval and visual prototypes and key geometric templates of corresponding family / type / variant entries in the visual model library to complete connector model determination. After obtaining the 3D definition provided by the 2D observation and vision model library of key geometric points, the pose of the connector relative to the camera is solved, and the stepwise coordinate transformation from the model coordinate system, camera coordinate system, workstation and / or robot arm base coordinate system is completed to obtain the target space coordinates and normal information of each pin or socket in the robot arm base coordinate system.

Images