Optical Instruments, Volume. 44, Issue 5, 42(2022)

Recognition of PCB solder joints and core wire ends

Hanlin LIU and Rongfu ZHANG*
Author Affiliations
  • School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China
  • show less
    Figures & Tables(33)
    Images of original solder joint and filtering effect
    Images of original core wire end and filtering effect
    HSV color space model
    HSV image of ROI single solder joint
    HSV image of ROI core wire end
    RGB single channel images
    Process of the soldering joint recognition algorithm
    Image of soldering joint segmentation result
    Implementation steps of mask image
    Mask image
    Mask results
    Schematic of contour center point
    Solder joint recognition results
    Flowchart of core wire end recognition algorithm
    Image multiplication
    Contrast enhancement effect
    Histogram equalization
    Morphological filtering
    Comparison of segmentation results
    Flowchart of connected domain analysis algorithm
    Core wire end recognition result
    Calibration results
    Algorithm comparison
    Solder joint recognition results of different cases
    Solder joint recognition results of two methods
    Schematic of solder joint distribution and naming method
    Core wire end recognition results of two methods
    • Table 1. HSV color space parameter list

      View table
      View in Article

      Table 1. HSV color space parameter list

      参数绿
      hminhmaxsminsmaxvminvmax0 180 0 255 0 46 0 180 0 43 46 220 0 180 0 30 221 255 0/156 10/180 43 255 46 255 11 25 43 255 46 255 26 34 43 255 46 255 35 77 43 255 46 255 78 99 43 255 46 255 100 124 43 255 46 255 125 155 43 255 46 255
    • Table 2. Calibration plate specification parameters

      View table
      View in Article

      Table 2. Calibration plate specification parameters

      名称外形尺寸图形尺寸直径中心距标定精度
      圆点标定板15 mm×15 mm6 mm×6 mm0.375 mm0.75 mm0.001 mm
    • Table 3. Comparison of coordinates between algorithm recognition and manual recognition of solder joint

      View table
      View in Article

      Table 3. Comparison of coordinates between algorithm recognition and manual recognition of solder joint

      焊点样本本文算法的焊点坐标/pixel人工识别的焊点坐标/pixel欧氏距离/pixel
      a(0,0) a(0,1) a(1,0) a(1,1) b(0,0) b(0,1) b(1,0) b(1,1) c(0,0) c(0,1) c(1,0) c(1,1) d(0,0) d(0,1) d(1,0) d(1,1) (380.00,528.00) (380.00,637.00) (512.00,528.00) (512.00,637.00) (1082.00,528.00) (1080.00,641.00) (1209.00,532.00) (1209.00,641.00) (380.00,1232.00) (380.00,1341.00) (506.00,1232.00) (506.00,1341.00) (1082.00,1232.00) (1077.00,1345.00) (1209.00,1232.00) (1208.00,1345.00) (382.00,530.00) (382.00,640.00) (516.00,530.00) (516.00,640.00) (1086.00,530.00) (1086.00,642.00) (1213.00,530.00) (1213.00,642.00) (383.00,1235.00) (383.00,1348.00) (511.00,1235.00) (511.00,1348.00) (1086.00,1234.00) (1086.00,1345.00) (1214.00,1236.00) (1214.00,1347.00) 2.83 3.60 4.47 5.00 4.47 6.08 4.47 4.12 4.24 7.62 5.84 8.60 4.47 9.00 5.39 6.32
    • Table 4. Comparison of measured values of center distance of solder joint circle (left and right)

      View table
      View in Article

      Table 4. Comparison of measured values of center distance of solder joint circle (left and right)

      焊点组号本文算法/μm人工测量/μm绝对值差值/μm误差/%
      a1 a2 b1 b2 c1 c2 d1 d2 450.00 450.00 433.16 439.77 429.55 429.55 432.95 446.59 456.82 456.82 432.95 432.95 436.36 436.36 436.43 436.43 6.82 6.82 0.21 6.82 6.81 6.81 3.48 10.16 1.49 1.49 0.05 1.58 1.56 1.56 0.80 2.33
    • Table 5. Comparison of measured values of center distance of solder joint circle (upper and lower)

      View table
      View in Article

      Table 5. Comparison of measured values of center distance of solder joint circle (upper and lower)

      焊点组号本文算法/μm人工实测/μm绝对值差值/μm误差/%
      a3 a4 b3 b4 c3 c4 d3 d4 371.60 371.60 385.23 371.60 371.60 371.60 446.97 385.40 375.00 375.00 381.82 381.82 385.23 385.23 378.41 378.41 3.40 3.40 3.41 10.22 13.63 13.63 68.56 6.99 0.90 0.91 0.89 2.68 3.54 3.54 18.12 1.85
    • Table 6. Comparison of coordinates between recognition algorithm and manual measurement of core wire end

      View table
      View in Article

      Table 6. Comparison of coordinates between recognition algorithm and manual measurement of core wire end

      芯线端头样本本文算法识别的像素坐标/pixel人工实测像素坐标/pixel欧氏距离/pixel
      1 2 3 4 5 6 7 8 9 10 (1789.00,534.00) (1241.00,767.00) (1788.00,355.00) (1718.00,502.00) (1842.00,510.00) (1874.00,582.00) (1751.00,534.00) (1094.00,1370.00) (1228.00,754.00) (1798.00,654.00) (1785.00,532.00) (1240.00,760.00) (1791.00,352.00) (1719.00,507.00) (1845.00,511.00) (1877.00,586.00) (1758.00,531.00) (1090.00,1373.00) (1230.00,750.00) (1802.00,650.00) 4.47 3.68 4.24 5.10 3.16 5.00 7.61 5.00 4.47 5.66
    Tools

    Get Citation

    Copy Citation Text

    Hanlin LIU, Rongfu ZHANG. Recognition of PCB solder joints and core wire ends[J]. Optical Instruments, 2022, 44(5): 42

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: APPLICATION TECHNOLOGY

    Received: Feb. 28, 2022

    Accepted: --

    Published Online: Nov. 14, 2022

    The Author Email: ZHANG Rongfu (zrf@usst.edu.cn)

    DOI:10.3969/j.issn.1005-5630.2022.005.006

    Topics