• DocumentCode
    2395333
  • Title

    VFD character display defect detection based on morphology and connected component labeling

  • Author

    Zhang, Caiyan ; Mu, Pingan ; Dai, Shuguang ; Wu, Minjie ; Yang, Jie

  • Author_Institution
    Sch. of Opt.-Electr. & Comput. Eng., Univ. of Shanghai for Sci. & Technol., Shanghai, China
  • fYear
    2012
  • fDate
    19-20 May 2012
  • Firstpage
    2008
  • Lastpage
    2011
  • Abstract
    According to the cursive code and leakage code defects of VFD characters display, This article proposes a new defect detection method. Firstly, we make use of pre-processing to original image. Then using binary morphological processing before recursive region growing method to mark the connected domins labeling. Finally calculate count of the VFD character display defects. According to the number of connected domains judge which defect is belong to. Experiments show that this detection method not only has nothing to do with the location of cursive code or leakage code defects, but also it is insensitive to the piston and tilt errors, This method has good accuracy and robustness.
  • Keywords
    automotive electronics; display devices; fault diagnosis; VFD character display defect detection; binary morphological processing; connected component labeling; cursive code defect; leakage code defect; recursive region growing method; vacuum fluorescent display; Filtering algorithms; Image segmentation; Labeling; Morphology; Noise; Radiation detectors; Shape; VFD character display; automatic detection; defect detection; image processing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Systems and Informatics (ICSAI), 2012 International Conference on
  • Conference_Location
    Yantai
  • Print_ISBN
    978-1-4673-0198-5
  • Type

    conf

  • DOI
    10.1109/ICSAI.2012.6223444
  • Filename
    6223444