• DocumentCode
    678092
  • Title

    Estimation of the Reflectance Distribution of Fallen Objects on Highways for Tunnel Lighting Management

  • Author

    Sakaguchi, So ; Mizutani, Daisuke ; Obama, Kengo ; Hirakawa, Satoshi ; Kaito, Kiyoyuki

  • Author_Institution
    Dept. of Civil Eng., Osaka Univ., Suita, Japan
  • fYear
    2013
  • fDate
    13-16 Oct. 2013
  • Firstpage
    4000
  • Lastpage
    4005
  • Abstract
    Fallen objects on highways are one of the direct or indirect factor which induce major accidents. Therefore, in management of highways, the improvement in visibility to fallen objects has been an important issue. And, the reflectance distribution of fallen objects and the tunnel lighting system greatly influence the improvement in the visibility of fallen objects. In this study, it is set the eventual goal that the establishment of the decision making of the tunnel lighting system, and as its basic study, it is precisely estimated the reflectance distribution of fallen objects. Specifically, it is represented the reflectance distribution by beta mixture model that can be considered multimodal. Mixing coefficient and unknown parameters of beta mixture model are estimated by the EM algorithm. Finally it is indicated the application example which used the survey datas, and verified of the validity of the model in this study.
  • Keywords
    decision making; lighting; object detection; reflectivity; road accidents; roads; tunnels; EM algorithm; beta mixture model; decision making; fallen objects visibility; highways; major accidents; mixing coefficient; reflectance distribution; tunnel lighting management; tunnel lighting system; Estimation; Lighting; Probability density function; Radiation detectors; Roads; Vehicles; beta mixture model; fallen objects on highways; the reflectance distribution;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Systems, Man, and Cybernetics (SMC), 2013 IEEE International Conference on
  • Conference_Location
    Manchester
  • Type

    conf

  • DOI
    10.1109/SMC.2013.683
  • Filename
    6722436