• DocumentCode
    125059
  • Title

    A novel way to process light-section image of wheel set outline based on B-spline function

  • Author

    Jing Li ; Xiaorong Gao ; Kai Yang

  • Author_Institution
    Nondestructive Testing Res. Center, Southwest Jiaotong Univ., Chengdu, China
  • fYear
    2014
  • fDate
    20-23 June 2014
  • Firstpage
    273
  • Lastpage
    277
  • Abstract
    In order to eliminate the shot noise and extract precise wheel contour line in online dynamic detection system for locomotive, threshold denoising method and smooth spline fitting are proposed in the paper. Furthermore, cubic B-spline interpolation is used to do subsequent processing on the principal curve. Before math processing on the curve, in advance the light-section image pixels are transformed into data points in rectangular coordinates to ensure higher precision of the wheel outline curve. Experimental results show that methods to process light-section image of wheel outline based on B-spline function is able to provide measurement data with higher precision when monitoring abrasion of key parts on wheel set.
  • Keywords
    interpolation; locomotives; nondestructive testing; shot noise; splines (mathematics); wheels; B-spline function; cubic B-spline interpolation; light-section image pixels; locomotive method; math processing; online dynamic detection system; rectangular coordinates; shot noise; smooth spline fitting; threshold denoising method; wheel contour line; wheel outline curve; wheel set outline; Fitting; Noise; Optical filters; Splines (mathematics); Vehicle dynamics; Wheels; Wiener filters; curve fitting; light-section method; spline function; wheel set outline;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Nondestructive Evaluation/Testing (FENDT), 2014 IEEE Far East Forum on
  • Conference_Location
    Chengdu
  • Print_ISBN
    978-1-4799-4731-7
  • Type

    conf

  • DOI
    10.1109/FENDT.2014.6928279
  • Filename
    6928279