• DocumentCode
    721069
  • Title

    An Automated Image Analysis Framework for Thermal Barrier Coating Porosity Measurement

  • Author

    Wei-Bang Chen ; Yongjin Lu ; Song Gao ; Chengcui Zhang ; Li, James ; Ogunbunmi, Olayinka S. ; Pradhan, Ligaj ; Ramsundar, Pallant ; Zimmerman, Ben

  • Author_Institution
    Dept. of Comput. Sci. & Eng., Virginia State Univ., VA, USA
  • fYear
    2015
  • fDate
    20-22 April 2015
  • Firstpage
    192
  • Lastpage
    195
  • Abstract
    Thermal barrier coating is a widely used advanced manufacturing technique. This paper introduces an image analysis based automated thermal barrier coating porosity measurement (TBCPM) framework. The proposed automated image analysis framework consists of two modules, including 1) top coat layer detection module, and 2) microstructure recognition and porosity measurement module. The first module is designed to automatically identify the top coat layer, a region of interest (ROI), in a thermal barrier coating image using a histogram-based approach. The second module recognizes the microstructures in the top coat layer using a local thresholding based image segmentation. The experimental results demonstrate that the porosity measurement produced from the proposed TBCPM framework is comparable to that of the domain experts.
  • Keywords
    image recognition; image segmentation; porosity; production engineering computing; thermal barrier coatings; automated image analysis; coat layer detection module; histogram-based approach; image segmentation; local thresholding; microstructure recognition; region of interest; thermal barrier coating porosity measurement; Biomedical measurement; Coatings; Image analysis; Image segmentation; Microstructure; Standards; Thermal analysis; image analysis; manufacturing automation; plasma sprayed coating; porosity; thermal barrier coating;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Multimedia Big Data (BigMM), 2015 IEEE International Conference on
  • Conference_Location
    Beijing
  • Print_ISBN
    978-1-4799-8687-3
  • Type

    conf

  • DOI
    10.1109/BigMM.2015.49
  • Filename
    7153876