Experiments in Fluids

, Volume 38, Issue 1, pp 21–32 | Cite as

Variational optical flow estimation for particle image velocimetry

Original Paper


We introduce a novel class of algorithms for evaluating PIV image pairs. The mathematical basis is a continuous variational formulation for globally estimating the optical flow vector fields over the whole image. This class of approaches has been known in the field of image processing and computer vision for more than two decades but apparently has not been applied to PIV image pairs so far. We pay particular attention to a multi-scale representation of the image data so as to cope with the quite specific signal structure of particle image pairs. The experimental evaluation shows that a prototypical variational approach competes in noisy real-world scenarios with three alternative approaches especially designed for PIV-sequence evaluation. We outline the potential of the variational method for further developments.


Particle Image Velocimetry Optical Flow Image Pair Digital Particle Image Velocimetry Interrogation Window Size 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. Barron J, Fleet D, Beauchemin S (1994) Performance of optical flow techniques. Int J Comput Vis 39:43–77Google Scholar
  2. Brandt A (1977) Multi-level adaptive solutions to boundary-value problems. Math Comput 31(138):333–390Google Scholar
  3. Brede M, Leder A, Westergaard CH (2003) Time-resolved PIV investigation of the separated shear layer in the transitional cylinder wake. In: Proceedings of 5th international symposium on PIV (PIV’03), Busan, KoreaGoogle Scholar
  4. Bruhn A, Weickert J, Feddern C, Kohlberger T, Schnörr C (2004) Variational optic flow computation in real-time. IEEE Trans Image Proc (in press)Google Scholar
  5. Chetverikov D (2003) Applying feature tracking to PIV. Int J Pattern Recognit Artif Intell 17(4):477–504Google Scholar
  6. Ciarlet PG (1978) The finite element method for elliptic problems. North-Holland, AmsterdamGoogle Scholar
  7. Hackbusch W (1985) Multigrid methods and applications. Springer, Berlin Heidelberg New YorkGoogle Scholar
  8. Hackbusch W (1993) Iterative solution of large sparse systems of equations. Springer, Berlin Heidelberg New YorkGoogle Scholar
  9. Horn BKP, Schunck BG (1981) Determining optical flow. Artif Intell 17:185–203CrossRefGoogle Scholar
  10. Kohlberger T, Schnörr C, Bruhn A, Weickert J (2004a) Domain decomposition for variational optical flow computation. IEEE Trans Image Proc (in press)Google Scholar
  11. Kohlberger T, Schnörr C, Bruhn A, Weickert J (2004b) Parallel variational motion estimation by domain decomposition and cluster computing. In: Pajdla T, Matas J (eds) Proc Eur Conf Comput Vis, Springer, Berlin Heidelberg New York, 3024:205–216Google Scholar
  12. Krishnamurthy R, Moulin P, Woods J (1996) Multiscale motion models for scalable video coding. In: Proceedings of IEEE international conference on image processing, Lausanne, SwitzerlandGoogle Scholar
  13. Okamoto K, Nishio S, Saga T, Kobayashi T (2000) Standard images for PIV. Measure Sci Technol 11:685–691CrossRefGoogle Scholar
  14. Quénot GM (1999) Performance evaluation of an optical flow technique applied to PIV using the VSJ standard images. In: Proceedings of 3rd international workshop on PIV, pp 579–584Google Scholar
  15. Quénot GM (1992) The orthogonal algorithm for optical flow detection using dynamic programming. In: Proceedings of international conference on acoustics, speech and signal, pp 249–252Google Scholar
  16. Quénot GM, Pakleza J, Kowalewski TA (1998) PIV with optical flow. Exp Fluids 25:177–189CrossRefGoogle Scholar
  17. Raffel M, Willert C, Kompenhans J (2001) Particle image velocimetry, 2nd edn. Springer, Berlin Heidelberg New YorkGoogle Scholar
  18. Scarano F (2002) Iterative image deformation methods in PIV. Measure Sci Technol 13:R1–R19CrossRefGoogle Scholar
  19. Schnörr C (1991) Determining optical flow for irregular domains by minimizing quadratic functionals of a certain class. Int J Comp Vis 6(1):25–38Google Scholar
  20. Schnörr C (1994) Segmentation of visual motion by minimizing convex non-quadratic functionals. In: Proceedings of 12th international conference on pattern recognition, Jerusalem, Israel, 9–13 OctGoogle Scholar
  21. Schnörr C (1999) Variational methods for adaptive image smoothing and segmentation. In: Jähne B, Haußecker H, Geißler P (eds) Handbook on computer vision and applications: Signal processing and pattern recognition. Academic, San Diego, 2:451–484Google Scholar
  22. Simoncelli EP (1993) Distributed representation and analysis of visual motion. PhD thesis, MITGoogle Scholar
  23. Weickert J, Schnörr C (2001) A theoretical framework for convex regularizers in pde–based computation of image motion. Int J Comput Vis 45(3):245–264CrossRefGoogle Scholar
  24. Weickert J, Schnörr C (2001) Variational optic flow computation with a spatio-temporal smoothness constraint. J Math Imaging Vis 14(3):245–255CrossRefGoogle Scholar
  25. Westergaard CH, Brede M, Leder A (2003) Time-space analysis of time resolved PIV data. In: Proceedings of 5th international symposium on PIV (PIV’03), Busan, KoreaGoogle Scholar

Copyright information

© Springer-Verlag 2004

Authors and Affiliations

  1. 1.Department of Mathematics and Computer ScienceComputer Vision, Graphics, and Pattern Recognition Group, University of Mannheim68131 MannheimGermany
  2. 2.Chair of Fluid Mechanics and AerodynamicsDarmstadt University of Technology64287 DarmstadtGermany

Personalised recommendations