Optics and Precision Engineering, Volume. 32, Issue 10, 1622(2024)

Computer motherboard assembly defect detection using parallel feature extraction and progressive feature fusion

Junying CHEN... Zhaoyang LI*, Hantao HUANG and Xuze DONG |Show fewer author(s)
Author Affiliations
  • School of Information and Control Engineering, Xi'an University of Architecture and Technology, Xi'an710055, China
  • show less
    Figures & Tables(15)
    Network architecture for defect detection in computer motherboard assembly
    Network structure of VTNet feature extraction network
    Network Structure of the VT-FasterBlock Feature Extraction Module
    Schematic diagram of the partial convolution structure
    Multiscale intersection attention asymptotic feature fusion networks
    Network Structure of Attention ASFF-2
    Network Structure of Attention ASFF-3
    Examples of types of surface assembly quality defects on computer motherboards
    Loss and mAP changes during the training process
    Computer motherboard assembly defect detection visualization test results
    Heat map of output scores for different structures introduced by the network
    • Table 1. Comparison of evaluation metrics for different defect detection algorithms

      View table
      View in Article

      Table 1. Comparison of evaluation metrics for different defect detection algorithms

      MethodBackbonemAP(@IoU=0.5)/%FPS/s
      Faster RCNNVGG1966.412
      YOLOv3DarkNet-5380.852
      YOLOv4CSPDarkNet5383.234
      YOLOxCSPDarkNet5389.151
      YOLOv5CSPDarkNet5390.056
      YOLOv7ELAN-Net93.319
      MobileNetV3-YOLOv423MobileNetV387.163
      Swin-Transformer-YOLOx24Swin-Transformer93.114
      OursVTNet94.625
    • Table 2. Results of ablation experiments with the proposed method

      View table
      View in Article

      Table 2. Results of ablation experiments with the proposed method

      MethodsmAP(@IoU=0.5)/%AP@(IoU=0.5)/%
      1234567891011
      YOLOv590.0194.0099.6699.4898.6098.6689.8540.7493.4093.4283.5198.80
      YOLOv5+改进191.7495.3298.2999.2898.4997.1086.9262.6994.2294.8982.7698.93
      YOLOv5+改进291.9595.4099.4198.5397.6394.6179.9084.7992.3994.2575.2698.96
      YOLOv5+改进1+改进294.6396.4198.4898.8398.1998.4688.5884.3899.7196.2282.6599.05
    • Table 3. Experimental results of different feature fusion networks

      View table
      View in Article

      Table 3. Experimental results of different feature fusion networks

      MethodsmAP(@IoU=0.5)/%FPSAP@(IoU=0.5)/%
      1234567891011
      YOLOv5+PANet90.015694.0099.6699.4898.6098.6689.8540.7493.4093.4283.5198.80
      YOLOv5+AFPN2091.084392.9695.9698.9098.4296.3688.3768.3791.5492.8379.5398.94
      YOLOv5+Ours91.952595.4099.4198.5397.6394.6179.9084.7992.3994.2575.2698.96
    • Table 4. Experimental results of different feature fusion networks

      View table
      View in Article

      Table 4. Experimental results of different feature fusion networks

      MethodsmAP(@IoU=0.5)/%FPSAP@(IoU=0.5)/%
      1234567891011
      YOLOv5-ResNet88.771795.0198.4999.2781.0295.5292.9646.6893.5196.9578.9898.19
      YOLOv5-CSPDarkNet90.015694.0099.6699.4898.6098.6689.8540.7493.4093.4283.5198.80
      YOLOv5-MobileViT89.614193.1798.6698.9991.4095.2393.0254.2193.7097.2574.7895.34
      YOLOv5-Swin-Transformer90.391693.8096.2997.0691.0595.2991.4158.2993.7194.2686.5696.51
      YOLOv5+Ours91.743395.3298.2999.2898.4997.1086.9262.6994.2294.8982.7698.93
    Tools

    Get Citation

    Copy Citation Text

    Junying CHEN, Zhaoyang LI, Hantao HUANG, Xuze DONG. Computer motherboard assembly defect detection using parallel feature extraction and progressive feature fusion[J]. Optics and Precision Engineering, 2024, 32(10): 1622

    Download Citation

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

    Category:

    Received: Dec. 8, 2023

    Accepted: --

    Published Online: Jul. 8, 2024

    The Author Email: LI Zhaoyang (nicholas@xauat.edu.cn)

    DOI:10.37188/OPE.20243210.1622

    Topics