Laser & Optoelectronics Progress, Volume. 56, Issue 9, 091501(2019)

Machine-Vision Based Defect Detection Algorithm for Packaging Bags

Dan Li*, Guojun Bai, Yuanyuan Jin, and Yan Tong
Author Affiliations
  • Department of Information and Control Engineering, Shenyang Urban Construction University, Shenyang, Liaoning 110167, China
  • show less
    Figures & Tables(13)
    Architectural diagram of machine-vision based detection system
    Flow chart of defect detection algorithm
    Binary images. (a) T1 threshold image; (b) T2 threshold image
    Picture of packing bag
    Platform for experimental testing
    Standard setting module
    Location setting module
    Partial detection results of defect classification. (a) Qualified image; (b) continuous bag (over length); (c) continuous bag (over width); (d) motion of packaging layout; (e) dimension error; (f) foreign matter on packages
    • Table 1. Feature and defect matching

      View table

      Table 1. Feature and defect matching

      No.ConditionDefect classification
      1L>Lup or W>WupDefect 1: continuous bag
      2L>Lup or L<Llow or W>Wup or W<WlowDefect 2: dimension error(over length or over width)
      3θ<θTDefect 3: foreign matteron packages
      4OMDefect 4: motion ofpackaging layout
    • Table 2. Confusion matrix

      View table

      Table 2. Confusion matrix

      ClassificationDetection
      DefectnumberQualifiednumber
      ActualDefect numberPTNF
      Qualified numberPFNT
    • Table 3. Confusion matrices for different detection methods

      View table

      Table 3. Confusion matrices for different detection methods

      CategoryProposedmethodTemplatematchingManualdetection
      DefectQualifiedDefectQualifiedDefectQualified
      Defect19551782218614
      Qualified62942527516284
    • Table 4. True positive rates, true negative rates, and accuracy of different detection methods

      View table

      Table 4. True positive rates, true negative rates, and accuracy of different detection methods

      MethodTrue positiverate /%True negativerate /%Accuracy /%
      Proposed method97.59897.8
      Template matching8991.790.6
      Manual detection9394.794
    • Table 5. Test of classification results

      View table

      Table 5. Test of classification results

      No.DefecttypeSampleSuccessnumberMissingnumberWrongnumberMissingrate /%Errorrate /%Positiverate /%
      1Continuous bag2001990100.599.5
      2Dimension error2001990100.599.5
      3Foreign matter on packages200195140.52.097.5
      4Motion of packaging layout2001970301.598.5
      Total800790190.1251.12598.75
    Tools

    Get Citation

    Copy Citation Text

    Dan Li, Guojun Bai, Yuanyuan Jin, Yan Tong. Machine-Vision Based Defect Detection Algorithm for Packaging Bags[J]. Laser & Optoelectronics Progress, 2019, 56(9): 091501

    Download Citation

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

    Category: Machine Vision

    Received: Oct. 10, 2018

    Accepted: Nov. 22, 2018

    Published Online: Jul. 5, 2019

    The Author Email: Dan Li (247573549@qq.com)

    DOI:10.3788/LOP56.091501

    Topics