• DocumentCode
    1979780
  • Title

    The vehicle speed measuring and precision analysis of video shot by moved camera based on 2D DLT algorithm

  • Author

    Gang Huang ; Chaozhong Wu ; Nengchao Lyu ; Pingfan Li

  • Author_Institution
    Intell. Transp. Syst. Res. Center, Wuhan Univ. of Technol., Wuhan, China
  • fYear
    2015
  • fDate
    25-28 June 2015
  • Firstpage
    111
  • Lastpage
    116
  • Abstract
    To solve the problem that the Direct Linear Transformation (DLT) algorithm cannot be used in the background moving scene, a novel transforming method based on 2D DLT is proposed. The method transforms the feature points in different scenes to the same object space coordinate. Firstly the precision of 2D DLT algorithm has been verified in high-precision 3-D digital camera calibration field. Experiments have been carried out on the platform of miniature traffic scene to calculate errors of the algorithm caused by movements of the camera. The results of the experiments indicated that the errors are all less than 2.5%. Although the errors present a rising trend, the limitation of actual traffic scene makes the proposed method available. The researching results can be used as an aid for speed measuring of road accident identified, and it can provide a scientific basis for the error estimation of the speed identification based on surveillance video which the backgrounds are changing.
  • Keywords
    image motion analysis; road accidents; road safety; road traffic; road vehicles; velocity measurement; video surveillance; 2D DLT algorithm; direct linear transformation; high-precision 3D digital camera calibration; road accident; surveillance video; vehicle speed measurement; Accidents; Calibration; Cameras; Measurement uncertainty; Safety; Vehicles; Direct Linear Transformation; Error analysis; Moved scene; Road traffic safety; Speed measurement; Traffic accident identification;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Transportation Information and Safety (ICTIS), 2015 International Conference on
  • Conference_Location
    Wuhan
  • Print_ISBN
    978-1-4799-8693-4
  • Type

    conf

  • DOI
    10.1109/ICTIS.2015.7232203
  • Filename
    7232203