Advertisement

Paladyn

, Volume 1, Issue 2, pp 141–146 | Cite as

Improving the performance of double-plane stereo vision system calibration, using a virtual plane

  • Paweł Rotter
  • Witold Byrski
  • Michał Dajda
  • Grzegorz Łojek
Research Article
  • 30 Downloads

Abstract

In the double-plane method for stereo vision system calibration, the correspondence between screen coordinates and location in 3D space is calculated based on four plane-to-plane transformations; there are two planes of the calibration pattern and two cameras. The method is intuitive, and easy to implement, but, the main disadvantage is ill-conditioning for some spatial locations. In this paper we propose a method which exploits the third plane which physically does not belong to the calibration pattern, but can be calculated from the set of reference points. Our algorithm uses a combination of three calibration planes, with weights which depend on screen coordinates of the point of interest; a pair of planes which could cause numerical errors receives small weights and have practically no influence on the final results. We analyse errors, and their distribution in 3D space, for the basic and the improved algorithm. Experiments demonstrate high accuracy and reliability of our method compared to the basic version; root mean square error and maximum error, are reduced by factors of 4 and 20 respectively.

Keywords

stereo vision camera calibration double-plane method 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    A. Bhatti (Ed.) Stereo Vision InTech 2008. Free access at: http://intechweb.org/book.php?id=87.
  2. [2]
    B. Cyganek, J.P. Siebert An Introduction to 3D Computer Vision Techniques and Algorithms. Wiley 2009.Google Scholar
  3. [3]
    V. Drenk, F. Hildebrand, M. Kindler & D. Kliche (1999). A 3D video technique for analysis of swimming in a flume. In R.H. Sanders, & B.J. Gibson (eds.), Scientific Proceedings of the XVII International Symposiumon Biomechanics in Sports (pp. 361–364). Perth, Australia: Edith-Cowan University.Google Scholar
  4. [4]
    O. Faugeras Three-Dimensional Computer Vision: A Geometric Viewpoint, MIT Press, 1993.Google Scholar
  5. [5]
    S. Fuksa, W. Byrski Czteropunktowa metoda identyfikacji transformacji stereowizyjnej (Four-pointidentification method of stereovision transformation) Automatyka 2005, 9(3).Google Scholar
  6. [6]
    R. Hartley, A. Zisserman Multiple View Geometry in Computer Vision. Cambridge University Press, 2003.Google Scholar
  7. [7]
    R. Klette, K. Schluns, A. Koschan Computer Vision: Three-Dimensional Data from Images. Springer-Verlag, 1998.Google Scholar
  8. [8]
    P.F. Sturm, S.J. Maybank On Plane-Based Camera Calibration: A General Algorithm, Singularities, Applications, CVPR, vol. 1, pp.1432, 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’99) — Volume 1, 1999.Google Scholar
  9. [9]
    A.V. Trucco Introductory Techniques for 3-D Computer Vision, Prentice Hall 1998.Google Scholar
  10. [10]
    Z. Zhang A flexible new technique for camera calibration. IEEE Transactions on PAMI, Volume 22,Issue 11, Nov 2000, pp. 1330–1334.Google Scholar
  11. [11]
    Z.Y. Zhang Flexible Camera Calibration by Viewing a Plane from Unknown Orientations, ICCV99(666–673), 1999. IEEE DOI Link BibRef 9900.Google Scholar

Copyright information

© © Versita Warsaw and Springer-Verlag Wien 2010

Authors and Affiliations

  • Paweł Rotter
    • 1
  • Witold Byrski
    • 1
  • Michał Dajda
    • 1
  • Grzegorz Łojek
    • 1
  1. 1.Automatics DepartmentAGH - University of Science and TechnologyKrakówPoland

Personalised recommendations