Fast and Accurate Vanishing Point Estimation on Structured Roads

  • Thomas WernerEmail author
  • Stefan Eickeler
Conference paper
Part of the Lecture Notes in Mobility book series (LNMOB)


We propose a method for estimating the vanishing point of structured roads directly in the image plane using the parallel nature of road markings as well as intelligent preprocessing and data reduction steps. The resulting vanishing point enables estimating the image to world projection, which then is used to perform subsequent tasks such as object detection. The major advantages of the proposed method are modest computational requirements as well as independence of the used camera model and without a calibration phase.


Fast vanishing point estimation Real time Camera independent Automotive Inverse perspective mapping Bird’s eye view Structured roads Road marking Multiresolution box filter 


  1. Bertozz M, Broggi A, Fascioli A (1998) Stereo inverse perspective mapping: theory and applications. Image Vis Comput 16(8):585–590CrossRefGoogle Scholar
  2. Borkar A, Hayes M, Smith M, Pankanti S (2009) A layered approach to robust lane detection at night. In: IEEE workshop on computational intelligence in vehicles and vehicular systems, March 2009, pp 51–57Google Scholar
  3. Deng J (2013) Fast lane detection based on the b-spline fitting. Int J Res Eng Technol:134–137Google Scholar
  4. Lin C, Wang M (2010) Topview transform model for the vehicle parking assistance system. In: International computer symposium, December 2010, pp 306–311Google Scholar
  5. Moghadam P, Starzyk J, Sardha W (2012) Fast vanishing-point detection in unstructured environments. IEEE Trans Image Process 21(1):425–430MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  1. 1.Fraunhofer Institute for Intelligent Analysis and Information SystemsSchloss Birlinghoven, Sankt AugustinGermany

Personalised recommendations