Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Laser & Optoelectronics Progress, Volume. 57, Issue 10, 101501(2020)

Aluminum Surface-Defect Detection Based on Multi-Task Deep Learning

Xiaohai Shen, Zehao Li, Min Li, Xiaolong Xu, and Xuewu Zhang*
Author Affiliations
  • College of Internet of Things Engineering, Hohai University, Changzhou, Jiangsu 213022, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (8)
    Equations (0)
    References (28)
    Cited By (8)
    Tools
    Paper Information
    Topics
    Figures & Tables(8)
    Parameters sharing in multi-task deep learning

    Fig. 1. Parameters sharing in multi-task deep learning

    Download full size
    Multi-task deep neural network model of aluminum defect detection

    Fig. 2. Multi-task deep neural network model of aluminum defect detection

    Download full size
    Adaptive multi-task loss layer

    Fig. 3. Adaptive multi-task loss layer

    Download full size
    Aluminum defect image and ground truth. (a) Original image; (b) segmentation marked; (c) multi-label marked; (d) defect object marked

    Fig. 4. Aluminum defect image and ground truth. (a) Original image; (b) segmentation marked; (c) multi-label marked; (d) defect object marked

    Download full size
    Results of model output on test sets. (a) Original images; (b) segmentation; (c) defect detection; (d) defect multi-label detection

    Fig. 5. Results of model output on test sets. (a) Original images; (b) segmentation; (c) defect detection; (d) defect multi-label detection

    Download full size

    • Table 1. Performance of single task training and multi-task training on test sets

      View table

      Table 1. Performance of single task training and multi-task training on test sets

      TaskTask weightMIoU(MPA) /%hloss /%mAP /%
      Seg.Clas.Obj.
      Segmentation10098.59(99.57)--
      Classification010-5.24-
      Object detection001--71.46
      Seg. +Clas.√√98.43(99.41)4.13-
      Seg. + Obj.√√98.82(99.73)-74.72
      Clas. + Obj.√√-1.0175.97
      Seg. +Clas. + Obj.√√√98.57(99.48)0.9675.65

    • Table 2. Performance of several deep learning methods on test sets

      View table

      Table 2. Performance of several deep learning methods on test sets

      MethodMIoU (MPA) /%hloss /%mAP /%
      FCN-8s98.75(99.49)--
      U-Net97.24(98.80)--
      ResNet50-6.41-
      ResNet101-4.67-
      Faster RCNN--69.78
      RetinaNet--72.13
      Proposed98.57(99.48)0.9675.65

    • Table 3. Inference time statics on single task model and multi-task model

      View table

      Table 3. Inference time statics on single task model and multi-task model

      TaskAverageinference timeper image /msSum of singletask inferencetime /ms
      Segmentation14.960-
      Classification13.791-
      Object detection57.347-
      Seg. +Clas.15.43728.751
      Seg. + Obj.68.21272.307
      Clas. + Obj.66.02571.138
      Seg. +Clas. + Obj.69.43586.098
    Tools

    Get Citation

    Copy Citation Text

    Xiaohai Shen, Zehao Li, Min Li, Xiaolong Xu, Xuewu Zhang. Aluminum Surface-Defect Detection Based on Multi-Task Deep Learning[J]. Laser & Optoelectronics Progress, 2020, 57(10): 101501

    Download Citation

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

    Category: Machine Vision

    Received: Jul. 15, 2019

    Accepted: Oct. 18, 2019

    Published Online: May. 8, 2020

    The Author Email: Xuewu Zhang (lab_112@126.com)

    DOI:10.3788/LOP57.101501

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology