Connector surface defect detection method and system

Abstract

The invention mainly relates to the technical field of industrial detection, in order to improve the precision and reliability of connector surface defect detection, the invention provides a connector surface defect detection method and system, and the core idea of the method is as follows: respectively determining exposure time for acquiring metal surface image features for different types of metal as first exposure time; aiming at different types of plastics, respectively determining exposure time for acquiring plastic surface image features as second exposure time; according to the types of metal and plastic on the surface of the to-be-detected connector, obtaining a connector surface image by adopting corresponding first exposure time and second exposure time; acquiring a surface feature image of the connector based on the image acquired in the step S2; and detecting and analyzing the surface feature image of the connector, and accurately and reliably judging the position of the defect on the surface of the connector and the corresponding defect type.

Description

Technical Field

[0001] The present invention mainly relates to the field of industrial detection technology, and in particular to a connector surface defect detection method and system. Background Art

[0002] As an important part of electronic devices, connectors are usually made of metal and plastic materials. During the production process, due to the high reflectivity of the metal surface and the low reflectivity of the plastic base, when detecting surface defects of connectors based on traditional cameras, the metal surface is prone to overexposure, while the plastic surface may appear too dark. The contrast between the two makes it difficult for the camera to clearly present the details of the metal and plastic areas at the same time.

[0003] Existing technologies usually use automatic exposure control (AEC) and high dynamic range (HDR) technology to cope with these challenges. However, in the actual inspection of connectors, due to the different surface materials, there are still problems such as excessive reflection of the metal part and weak reflection of the plastic part, which affects the overall clarity of the image and thus affects the accuracy of connector surface defect detection. Therefore, a technical solution is urgently needed to solve the problem of metal reflection on the connector surface and weak reflection on the plastic base by optimizing exposure control, so as to achieve more accurate detection results of connector surface defects. Summary of the invention

[0004] Technical Problems to be Solved by the Invention

[0005] A connector surface defect detection method and system are provided, aiming to improve the accuracy and reliability of connector surface defect detection.

[0006] The technical solution adopted by the present invention to solve the above technical problems

[0007] In one aspect, the present invention provides a method for detecting surface defects of a connector, comprising the following steps:

[0008] S1: for different types of metals, respectively determine the exposure time for acquiring metal surface image features as the first exposure time, and for different types of plastics, respectively determine the exposure time for acquiring plastic surface image features as the second exposure time;

[0009] S2: acquiring a connector surface image by using a corresponding first exposure time and a second exposure time according to the types of metal and plastic on the connector surface to be detected;

[0010] S3: Acquire a surface feature image of the connector based on the image acquired in step S2;

[0011] S4: Detect and analyze the connector surface feature image to determine the location of defects on the connector surface and the corresponding defect type.

[0012] Furthermore, the metal surface image features in step S1 include texture, scratches and oxidation spots on the metal surface.

[0013] Furthermore, the plastic surface image features in step S1 include texture, cracks, bubbles and stains on the plastic surface.

[0014] Furthermore, step S2 further includes: acquiring an image of the interface between metal and plastic on the connector surface based on the set exposure parameters as a supplementary image to the connector surface image acquired using the corresponding first exposure time and second exposure time.

[0015] Furthermore, the exposure parameters include a set exposure time, an image shooting angle, and an auxiliary light source intensity used to shoot the image.

[0016] Furthermore, multiple groups of exposure parameters are set to obtain multiple images of the interface between metal and plastic on the connector surface as supplementary images of the connector surface obtained by using the corresponding first exposure time and second exposure time.

[0017] Furthermore, the step S3 of acquiring the connector surface feature image includes:

[0018] S31: performing image correction on the image acquired in step S2 by feature point registration or frequency domain-based phase correction, and performing pixel alignment on the corrected image;

[0019] S32: The pixel-aligned image is subjected to pixel feature fusion by weighted averaging, multi-scale fusion or HDR algorithm to achieve a balance between bright and dark areas;

[0020] S33: Optimize the brightness and contrast of the fused image.

[0021] Further, step S33 includes:

[0022] The histogram equalization method is used to enhance the overall contrast of the image;

[0023] Gamma correction is used to improve the brightness level of local areas of the image;

[0024] Adaptive contrast is used to enhance the image features of dark or bright areas in the image.

[0025] Furthermore, step S4 specifically includes:

[0026] S41: performing edge detection, threshold segmentation, and morphological operations on the connector surface feature image, segmenting the connector surface area and marking defective areas;

[0027] S42: Combine machine learning or deep learning algorithms to determine the location and type of the extracted defects.

[0028] On the other hand, based on the above-mentioned connector surface defect detection method, the present invention also provides a connector surface defect detection system, the system comprising an image acquisition module, an image processing module and a defect detection module;

[0029] The image acquisition module is used to acquire an image of the connector surface and an image of the interface between metal and plastic on the connector surface according to the first exposure time and the second exposure time;

[0030] The image processing module is used to obtain a connector surface feature image based on an image of the connector surface and an image of a junction between metal and plastic on the connector surface;

[0031] The defect detection module detects and analyzes the connector surface feature image to determine the defect type on the connector surface.

[0032] Beneficial Effects of the Invention

[0033] (1) The present invention discloses a method and system for detecting defects on the surface of a connector. The method and system first determine the exposure time for obtaining detailed image features of different metal and plastic surfaces, respectively. The first exposure time and the second exposure time are selected according to the material properties of the metal and plastic on the surface of the connector to be detected, respectively, to obtain the image features of the metal and plastic on the surface of the connector. The image is processed to obtain a connector surface feature image that can display the entire surface of the connector. The connector surface feature image is detected and analyzed to determine the location of defects on the surface of the connector and the corresponding defect type. The problem of excessive metal reflection on the surface of the connector and weak reflection on the plastic base can be effectively solved, thereby ensuring the detection accuracy and reliability of connector defects, thereby improving production efficiency and ensuring the normal and orderly progress of the production process.

[0034] (2) In the process of generating the surface feature image of the connector, the present invention also obtains multiple images of the junction between the metal part and the plastic part of the connector surface according to the set exposure parameters, as a supplementary image of the connector surface portrait obtained according to the first exposure time and the second exposure time, which is used to capture the detailed features of the connector surface and improve the reliability of the generated connector surface feature image. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a flow chart of a connector surface defect detection method according to the present invention;

[0036] Figure 2 This is a schematic diagram of a connector surface defect detection system according to the present invention. DETAILED DESCRIPTION

[0037] The specific embodiments of the present invention are described in detail below, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention. The components of the embodiments of the present invention generally described and shown in the drawings herein may be arranged and designed in a variety of different configurations.

[0038] like Figure 1 As shown, a connector surface defect detection method described in the present invention specifically includes the following steps.

[0039] S1: According to the metal types commonly used in connectors, the exposure time for obtaining the corresponding metal surface image features is determined for different types of metals as the first exposure time. The specific duration of the first exposure time is set according to the specific reflection characteristics of different metal surfaces. The core is to effectively suppress the overexposure of the corresponding metal during the image acquisition process, so that the texture, scratches, oxidation spots and other details of the metal surface can be clearly presented.

[0040] According to the types of plastics commonly used in connectors, the exposure time for obtaining the image features of the plastic surface is determined for each type of plastic as the second exposure time. The specific duration of the second exposure time is set in accordance with the principle of being able to obtain detailed features such as texture, cracks, bubbles and stains on the plastic surface.

[0041] S2: Acquire the connector surface image using the corresponding first exposure time and second exposure time according to the types of metal and plastic on the connector surface to be inspected; in the image acquired according to the first exposure time, the information of the metal area on the connector surface can be completely retained, while the dark details of the plastic area may be missing due to weak reflection; in the image acquired according to the second exposure time, although the metal area may be overexposed, the information of the plastic area can be fully recorded.

[0042] Preferably, in this embodiment, an image of the junction area between the metal area and the plastic area on the connector surface is obtained based on the set exposure parameters. By setting different exposure parameters (such as exposure time, camera angle, auxiliary light source intensity, etc.), images with more exposure gradients are obtained to capture the details of the junction between metal and plastic or special structural areas, ensuring the integrity of the full surface information.

[0043] S3: Synthesize and fuse images taken from multiple exposures to generate a high dynamic range image containing metal and plastic details, providing reliable input for subsequent inspection.

[0044] S21. Image alignment: Since there may be camera shake or slight movement of the connector position during multiple exposures, each image needs to be accurately aligned. Use feature point registration (such as SIFT or ORB algorithm) or frequency domain-based phase correlation method for image correction to ensure that all images are aligned at the pixel level to avoid ghosting or blurring after fusion.

[0045] S22. Image fusion: The aligned images are fused with information, and weighted average, multi-scale fusion or HDR (high dynamic range) synthesis algorithm is used to achieve a balance of details in bright and dark areas. Specifically, weighted average fusion can give priority to the bright information of short-exposure images and the dark information of long-exposure images. Multi-scale fusion is based on the pyramid algorithm to optimize image details at different scales. HDR synthesis unifies multiple images into a high dynamic range image through radiometric curves or tone mapping technology.

[0046] S23, brightness and contrast adjustment: optimize the brightness and contrast of the fused image to make the details of the metal and plastic areas clearer and more discernible; specifically, the histogram equalization method can be used to enhance the overall contrast of the image, gamma correction can be used to enhance the brightness level of the local area, and adaptive contrast enhancement (such as CLAHE) can be used to further highlight the details of the dark or bright areas.

[0047] S3: Detect and analyze the connector surface feature image to determine the location of defects on the connector surface and the corresponding defect type.

[0048] S31. Defect extraction: Analyze high dynamic range images through image processing algorithms to extract possible defect information. Use edge detection (such as Canny or Sobel operators), threshold segmentation (based on global or local adaptive methods) and morphological operations (such as corrosion and expansion) to segment the connector surface area and mark suspected defect areas.

[0049] S32. Defect classification and judgment: Combine machine learning or deep learning algorithms to classify and judge the extracted defects, use pre-trained models to identify defect types such as scratches, cracks, bubbles, stains, etc., and output the location of the defects.

[0050] like Figure 2 As shown, a connector surface defect detection system according to the present invention comprises an image acquisition module, an image processing module and a defect detection module;

[0051] The image acquisition module is used to acquire an image of the connector surface and an image of the interface between metal and plastic on the connector surface according to the first exposure time and the second exposure time;

[0052] The image processing module is used to obtain a connector surface feature image based on an image of the connector surface and an image of a junction between metal and plastic on the connector surface;

[0053] The defect detection module detects and analyzes the connector surface feature image to determine the defect type on the connector surface.

[0054] The connector surface defect detection system is installed in the automated production line of the connector. The image acquisition module performs short-exposure photography of the metal surface according to the set first exposure time, and performs long-exposure photography of the plastic parts such as the base according to the set second exposure time, and then obtains images of special areas such as the junction of metal and plastic according to the set exposure parameters based on the type of connector and other characteristics; the image processing module generates a connector surface feature image that can fully display the details of the metal and plastic areas based on the image collected by the image acquisition module, and the defect detection module detects and analyzes the generated connector surface feature image according to the set defect detection algorithm, automatically identifies whether there are defects such as textures, scratches, oxidation spots, cracks, bubbles and stains on various parts of the connector, and the location of the corresponding defects, and detects them according to the set defect detection algorithm. Figure 1 As shown, the qualified or unqualified judgment result of the connector production is further output based on the defect detection result. If qualified, the product is retained. If unqualified, the defect detection result is fed back to the production line in the form of a signal or data to achieve real-time defect removal or alarm.

[0055] The present invention is not limited to the above-mentioned optional implementation modes. Anyone can derive other various forms of products under the inspiration of the present invention. However, no matter what changes are made in the shape or structure, all technical solutions that fall within the scope defined by the claims of the present invention fall within the protection scope of the present invention.

Claims

1. A method for detecting surface defects of a connector, characterized in that: The following steps are involved: S1: for different types of metals, respectively determine the exposure time for acquiring metal surface image features as the first exposure time, and for different types of plastics, respectively determine the exposure time for acquiring plastic surface image features as the second exposure time; S2: acquiring a connector surface image by using a corresponding first exposure time and a second exposure time according to the types of metal and plastic on the connector surface to be detected; S3: Acquire a surface feature image of the connector based on the image acquired in step S2; S4: Detect and analyze the connector surface feature image to determine the location of defects on the connector surface and the corresponding defect type.

2. A connector surface defect detection method according to claim 1, characterized in that: The metal surface image features in step S1 include texture, scratches and oxidation spots on the metal surface.

3. A connector surface defect detection method according to claim 1, characterized in that: The plastic surface image features in step S1 include texture, cracks, bubbles and stains on the plastic surface.

4. A connector surface defect detection method according to claim 1, characterized in that: Step S2 also includes: acquiring an image of the interface between metal and plastic on the connector surface based on the set exposure parameters as a supplementary image to the connector surface image acquired using the corresponding first exposure time and second exposure time.

5. A connector surface defect detection method according to claim 4, characterized in that: The exposure parameters include the set exposure time, the image shooting angle and the auxiliary light source intensity used for shooting the image.

6. A connector surface defect detection method according to claim 5, characterized in that: A plurality of groups of exposure parameters are set to obtain a plurality of images of the interface between metal and plastic on the connector surface as supplementary images of the connector surface images obtained by using the corresponding first exposure time and second exposure time.

7. A connector surface defect detection method according to any one of claims 1 to 6, characterized in that: The step S3 of acquiring the connector surface feature image comprises: S31: performing image correction on the image acquired in step S2 by feature point registration or frequency domain-based phase correction, and performing pixel alignment on the corrected image; S32: The pixel-aligned image is subjected to pixel feature fusion by weighted averaging, multi-scale fusion or HDR algorithm to achieve a balance between bright and dark areas; S33: Optimize the brightness and contrast of the fused image.

8. A connector surface defect detection method according to claim 7, characterized in that: Step S33 includes: The histogram equalization method is used to enhance the overall contrast of the image; Gamma correction is used to improve the brightness level of local areas of the image; Adaptive contrast is used to enhance the image features of dark or bright areas in the image.

9. A connector surface defect detection method according to claim 1, characterized in that: Step S4 specifically includes: S41: performing edge detection, threshold segmentation, and morphological operations on the connector surface feature image, segmenting the connector surface area and marking defective areas; S42: Combine machine learning or deep learning algorithms to determine the location and type of the extracted defects.

10. A connector surface defect detection system, used to implement the connector surface defect detection method according to any one of claims 1 to 9, characterized in that: It includes an image acquisition module, an image processing module and a defect detection module; The image acquisition module is used to acquire an image of the connector surface and an image of the interface between metal and plastic on the connector surface according to the first exposure time and the second exposure time; The image processing module is used to obtain a connector surface feature image based on an image of the connector surface and an image of a junction between metal and plastic on the connector surface; The defect detection module detects and analyzes the connector surface feature image to determine the defect type on the connector surface.

Images