• DocumentCode
    1335681
  • Title

    High-Speed Image Processing Algorithms for Real-Time Detection of MARFEs on JET

  • Author

    De Albuquerque, Márcio Portes ; De Albuquerque, Marcelo Portes ; Chacon, Germano T. ; de Faria, E.L. ; Murari, Andrea

  • Author_Institution
    Centro Brasileiro de Pesquisas Fisicas (CBPF), Rio de Janeiro, Brazil
  • Volume
    40
  • Issue
    12
  • fYear
    2012
  • Firstpage
    3485
  • Lastpage
    3492
  • Abstract
    This paper presents a high-speed image processing algorithm applied to multifaceted asymmetric radiation from the edge detection on Joint European Torus. The algorithm has been developed in C/C++ language using the OpenCV library for image processing and the LibSVM library for pattern classification and has been tested on a dedicated Intel 64-bit Linux computing platform. The code implemented achieves at the same time high accuracy and good run-time performance. The final version presents a correct detection rate of 93.3% and an average image processing rate of 650 frames per second. A complete analysis of each image processing module is presented to illustrate the overall characteristics and performance of the algorithm.
  • Keywords
    C++ language; Linux; Tokamak devices; edge detection; image classification; plasma diagnostics; plasma toroidal confinement; 64-bit Linux computing platform; C language; C++ language; JET; Joint European Torus; LibSVM library; MARFE; OpenCV library; average image processing rate; detection rate; edge detection; high-speed image processing algorithm; image processing module; multifaceted asymmetric radiation; pattern classification; real-time detection; Algorithm design and analysis; Cameras; Image processing; Libraries; Magnetic confinement; Random access memory; Support vector machines; Image processing; MARFE identification; multifaceted asymmetric radiation from the edge (MARFE) detection; real-time MARFE processing;
  • fLanguage
    English
  • Journal_Title
    Plasma Science, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0093-3813
  • Type

    jour

  • DOI
    10.1109/TPS.2012.2223235
  • Filename
    6353975