• DocumentCode
    2783314
  • Title

    Lane Detection Based on the Random Sample Consensus

  • Author

    Guo Keyou ; Li Na ; Zhang Mo

  • Author_Institution
    Sch. of Mater. & Mech. Eng., Beijing Technol. & Bus. Univ., Beijing, China
  • Volume
    3
  • fYear
    2011
  • fDate
    24-25 Sept. 2011
  • Firstpage
    38
  • Lastpage
    41
  • Abstract
    In order to improve real-time and robustness of the lane detection and get more ideal lane, in the image preprocessing, the filter is used in strengthening lane information of the binary image, reducing the noise and removing irrelevant information. The lane edge detection is by using Canny operator, then the corner detection method is used in getting the Image corners coordinates and finally using the RANSAC to circulation fit for corners, according to the optimal lanes parameters drawing lane. Through experiment of different scenes, this method can not only effectively rule out linear pixel interference of outside the road in multiple complex environments, but also quickly and accurately identify lane. This method improves the stability of the lane detection to a certain extent, which has good robust and real-time.
  • Keywords
    edge detection; filtering theory; image denoising; random processes; traffic engineering computing; Canny operator; RANSAC; binary image; corner detection method; image corners; image preprocessing; irrelevant information removal; lane edge detection; lane information; linear pixel interference; noise reduction; optimal lane parameter drawing lane; random sample consensus; Algorithm design and analysis; Computers; Educational institutions; Image edge detection; Navigation; Real time systems; Robustness; Corner detection; Edge detection; Lane detection; RANSAC;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information Technology, Computer Engineering and Management Sciences (ICM), 2011 International Conference on
  • Conference_Location
    Nanjing, Jiangsu
  • Print_ISBN
    978-1-4577-1419-1
  • Type

    conf

  • DOI
    10.1109/ICM.2011.93
  • Filename
    6113579