CN-122017383-A - Cable electrical property detection equipment and method based on visual recognition

CN122017383ACN 122017383 ACN122017383 ACN 122017383ACN-122017383-A

Abstract

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

Inventors

XING JUNHAO
GUO JIANHUA
LI PENGCHENG

Assignees

北京机械设备研究所

Dates

Publication Date: 20260512
Application Date: 20251218

Claims (10)

1. A visual identification-based cable electrical property detection apparatus, comprising: The mechanical arm module is used for installing the machine vision identification module and the probe test module and moving relative to the electric connector of the tested cable in space under a control instruction; the machine vision identification module is arranged on the mechanical arm module and is used for acquiring images of the electric connector, matching and matching the acquired images with a visual model library established in advance, identifying the model of the electric connector and acquiring the space coordinates and normal information of each contact pin or jack; The probe testing module is arranged at the tail end of the mechanical arm module, is used for moving to the space coordinate under the drive of the mechanical arm module, is in contact with a corresponding contact pin or jack, and is used for applying a testing signal to a measured point position and collecting a response signal when the probe testing module is in electrical contact with the measured point position so as to execute at least one electrical performance test of a passage, an insulation resistance and a withstand voltage and outputting a testing result; The industrial personal computer is in communication connection with the mechanical arm module, the machine vision identification module and the probe test module and is used for running test system software to control vision identification, mechanical arm movement, electrical property test and test data recording and display; The probe end of the probe test module is directly contacted with an electric connector of a tested cable and forms a test loop with a circuit for electric performance test, and a prefabricated tool cable is not arranged in the middle to serve as an intermediate connecting piece, so that electric performance detection of cables of different types is completed under the same equipment structure.
2. The apparatus of claim 1, wherein the machine vision recognition module comprises an industrial camera, a light source, and an image processing unit configured to: after the internal camera parameter calibration and the hand-eye calibration of a camera-mechanical arm and/or a station coordinate system are finished, controlling the industrial camera to acquire a target image according to given exposure, gain, polarization and light field parameters; Denoising, distortion correction, brightness normalization and highlight/reflection suppression are performed on the acquired target image, and a standardized image frame for identification is obtained.
3. The device of claim 1, wherein the image processing unit is further configured to: Reasoning the standardized image based on the trained target detection network, and outputting a candidate region containing the connector appearance and an example mask; Performing instance segmentation or contour extraction based on edge detection and morphological operation in the candidate region to obtain an outer contour and an inner cavity contour of the connector, and identifying at least one key element of a positioning key, a buckle or a guide surface; Parametric fitting is performed on the duplicate hole/pin array and pins or receptacles are numbered in a given order.
4. The apparatus of claim 1, wherein the library of visual models is constructed by: Image acquisition and camera calibration are carried out on the connector samples, and standardized image frames and acquisition metadata are obtained; Obtaining the outer contour and the inner cavity contour of each connector by using a candidate region detection and contour extraction method, and identifying key geometric elements of a positioning key, a guide surface and a buckle; Numbering each contact pin or jack, determining nominal three-dimensional coordinates and normal information of each point location under a model coordinate system, and establishing mapping with a manufacturer naming table; And (3) structurally packaging and storing the original image, the candidate region, the profile data, the numbering table, the array parameters, the coordinate system definition and the calibration and acquisition metadata into a database, and establishing an index according to the connector family/type/variant to form a visual model entry.
5. The apparatus of claim 1, wherein the machine vision recognition module, after acquiring the two-dimensional observation of the keypoint, is further configured to: Calling a visual model entry corresponding to the identified connector family/type/variant, and acquiring key geometric points of the connector under a model coordinate system and three-dimensional nominal coordinates of each contact pin/jack; Solving the pose of the connector relative to the camera according to the correspondence between the internal parameters of the camera and the key points; and step-by-step coordinate transformation of the model coordinate system, the camera coordinate system, the station coordinate system and/or the mechanical arm base coordinate system is completed, the space coordinate and the normal information of each contact pin or jack under the mechanical arm base coordinate system are obtained, and the space coordinate and the normal information are used as input of mechanical arm path planning.
6. The apparatus of claim 1, further comprising a clamping module, the clamping module comprising: an adjustable guide rail that is arrangeable in a length direction; The movable clamping blocks are arranged on the guide rail, and the positions of the movable clamping blocks can be adjusted along the guide rail so as to adapt to the shapes of the electric connectors with different widths or lengths; the elastic cushion layer is arranged on the contact surface of the movable clamping block and the electric connector and is used for improving the clamping stability and reducing the local stress in the clamping process; And the clamping module is configured to form a flexible universal tool so as to finish clamping of various types of electric connectors through simple adjustment.
7. The apparatus of claim 1, wherein the probe test module comprises a probe module and an electrical performance test module; the probe module comprises at least one probe and a probe mounting seat, wherein the probe mounting seat is rigidly connected with the tail end of the mechanical arm, and the probe extends out of the probe mounting seat along a preset direction; The electrical performance testing module is electrically connected with the probe module and is used for applying a testing signal to the probe and collecting a response signal; and when the control mechanical arm module drives the probe test module to move to the target space coordinate, the industrial personal computer is configured to set an included 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 electric contact with the contact pin or the jack.
8. The apparatus of claim 1, further comprising a safety device comprising a safety grating and a scram button, wherein when the safety grating detects a person entering a detection area and/or the scram button is triggered, the industrial personal computer issues a stop instruction to the robot module and the probe test module through an apparatus control module of the test system software to stop the robot movement and close the high voltage output of the probe test module, and the current detection state is recorded by a data storage and trace module.
9. A method for detecting electrical properties of a cable using the apparatus of any one of claims 1 to 8, comprising: 1) Fixing the electric connector of the tested cable to a clamping module and/or a fixing station; 2) The method comprises the steps of image acquisition and preprocessing, namely moving a machine vision identification module to a preset shooting position through a mechanical arm module, acquiring a target image of the electric connector through an industrial camera, and carrying out denoising, distortion correction, brightness normalization and highlight/reflection inhibition processing on the target image to obtain a standardized image frame; 3) The image recognition and point location analysis steps comprise the steps of detecting candidate areas containing the electric connectors from the standardized image frames based on a target detection network and example segmentation/contour extraction, recognizing positioning keys, buckles, guide surfaces and repeated holes/needle arrays, calling a visual model library to execute matching pairs, and determining the model of the tested electric connectors and the three-dimensional space coordinates and normal information of each contact pin or jack; 4) Planning a motion track of the mechanical arm and a contact gesture of the probe by the industrial personal computer according to the three-dimensional space coordinates and the normal information, and controlling the mechanical arm module to drive the probe testing module to sequentially move to each point to be tested and contact with the point to be tested; 5) And the electrical performance test and result judgment step is that after the probe test module is reliably contacted with each point to be tested, the probe test module executes at least one electrical performance test of a passage, an insulation resistance and a withstand voltage, the test result is returned to the industrial personal computer, the industrial personal computer judges the test result according to a preset qualified criterion, and the test data recording and display are completed.
10. The method of claim 9, wherein in the image recognition and point location resolution step, further comprising: extracting multi-scale appearance characteristics and geometric key point responses from the candidate areas, and performing similarity calculation on visual prototypes and key geometric templates of corresponding family/type/variant items in the visual model library through measurement retrieval to finish connector model judgment; after obtaining the two-dimensional observation of the key geometric points and the three-dimensional definition provided by the visual model library, solving the pose of the connector relative to the camera, and completing the step-by-step coordinate transformation of a model coordinate system, a camera coordinate system, a station and/or a mechanical arm base coordinate system so as to obtain the target space coordinate and normal information of each contact pin or jack under the mechanical arm base coordinate system.

Description

Cable electrical property detection equipment and method based on visual recognition Technical Field The disclosure relates to the technical field of cable detection, in particular to a device and a method for detecting electrical performance of a cable based on visual identification. Background After the production of the electric device is finished, the electric performance items such as a passage, an insulation resistor, a withstand voltage and the like are usually required to be detected, so that the product quality and the use safety are ensured. The current cable production is generally divided into batch production projects and development projects, wherein for the batch production projects, a mode of 'cable detection equipment and tool cables' is generally adopted, and high-efficiency electrical performance detection of batch cables is realized through cooperation of special tool cables manufactured in advance and a general test bench. Therefore, at present, handheld instruments such as a universal meter, an insulating meter, a voltage withstanding instrument and the like are generally adopted in the development stage, and related points of the cable connector are manually tested one by one according to the process file. This approach has the following disadvantages: 1. The efficiency is low, the test time is long, and the requirements of frequent modification and quick verification in the development stage are difficult to meet; 2. Relying on the experience of operators, the connector has multiple points and complex distribution, and the manual point-by-point contact is easy to cause missed detection, false detection or poor contact; 3. the consistency and traceability are poor, namely, manual operation is difficult to keep the complete consistency of each detection flow, the detection result is difficult to trace, and the quality problem analysis and responsibility division are not facilitated; 4. And if the efficiency is improved, a large number of special tool cables still need to be manufactured, and the investment is not matched with the income for developing small-batch projects. In summary, the prior art has not yet lacked a general detection device that does not need to separately manufacture a tooling cable for each cable, can automatically identify the cable connector point location through machine vision, and can automatically complete probe contact and electrical performance test by driving a mechanical arm, and has good generalization capability and data tracing function. It should be noted that the information disclosed in the above background section is only for enhancing understanding of the background of the present disclosure and thus may include information that does not constitute prior art known to those of ordinary skill in the art. Disclosure of Invention It is an object of the present disclosure to provide a cable electrical property detection apparatus and method based on visual recognition, which further overcome, at least in part, one or more of the problems due to the limitations and disadvantages of the related art. To achieve the above object, an aspect of the present invention provides a cable electrical property detection apparatus based on visual recognition, including: The mechanical arm module is used for installing the machine vision identification module and the probe test module and moving relative to the electric connector of the tested cable in space under a control instruction; the machine vision identification module is arranged on the mechanical arm module and is used for acquiring images of the electric connector, matching and matching the acquired images with a visual model library established in advance, identifying the model of the electric connector and acquiring the space coordinates and normal information of each contact pin or jack; The probe testing module is arranged at the tail end of the mechanical arm module, is used for moving to the space coordinate under the drive of the mechanical arm module, is in contact with a corresponding contact pin or jack, and is used for applying a testing signal to a measured point position and collecting a response signal when the probe testing module is in electrical contact with the measured point position so as to execute at least one electrical performance test of a passage, an insulation resistance and a withstand voltage and outputting a testing result; The industrial personal computer is in communication connection with the mechanical arm module, the machine vision identification module and the probe test module and is used for running test system software to control vision identification, mechanical arm movement, electrical property test and test data recording and display; The probe end of the probe test module is directly contacted with an electric connector of a tested cable and forms a test loop with a circuit for electric performance test, and a prefabricated tool cable is not arranged in the middle