System Architecture for Future Driver Assistance Based on Stereo Vision

  • Thomas Wehking
  • Alexander Würz-Wessel
  • Wolfgang Rosenstiel


The last two decades saw much research on computer vision in automotive environments. Since a few years the first and second generation of vision based driver assistance systems are available in luxury and middle-class vehicles. The offered functionality is limited to comfort tasks like road sign recognition, night view and lane keeping support. To develop more complex functions with a safety aspect like pedestrian protection, the required measurement data has to be more accurate. The scene depth information must be available and a robust model free approach is preferable.


  1. 1.
    Hartley R, Zisserman A (2003) Multiple view geometry in computer vision, 2nd edn. Cambridge University Press, Cambridge, p 237ffGoogle Scholar
  2. 2.
    Jähne B (2005) Digitale Bildverarbeitung, 6th edn. Springer, Berlin, p 423ffGoogle Scholar
  3. 3.
    Ken O’Neill (2004) Natürliche Strahlung verursacht Fehler in FPGA’s, elektronik industrie, vol 11, pp 52–53Google Scholar
  4. 4.
    Lorei M, Würz-Wessel A, Heger T (2008) 6D-Stereo ein robuster und schneller Ansatz zur Objektdetektion, VDI-Berichte 2038, pp 85–92Google Scholar
  5. 5.
    Wehking T, Würz-Wessel A, Rosenstiel W (2009) 6D-Stereo Video für moderne Fahrerassistenzsysteme. Workshop Fahrerassistenzsysteme, Löwenstein/Hößlinsülz, pp 20–28Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Thomas Wehking
    • 1
  • Alexander Würz-Wessel
    • 1
  • Wolfgang Rosenstiel
    • 2
  1. 1.Robert Bosch GmbHLeonbergGermany
  2. 2.Eberhard-Karls-UniversityTübingenGermany

Personalised recommendations