Optical Stokes flow estimation: an imaging-based control approach

被引:65
作者
Ruhnau, Paul [1 ]
Schnoerr, Christoph [1 ]
机构
[1] Univ Mannheim, Dept Math & Comp Sci, Comp Vis Graph & Pattern Recognit Grp, D-68131 Mannheim, Germany
关键词
Particle Image Velocimetry; Stokes Equation; Optical Flow; Image Pair; Adjoint Equation;
D O I
10.1007/s00348-006-0220-z
中图分类号
TH [机械、仪表工业];
学科分类号
0802 ;
摘要
We present an approach to particle image velocimetry based on optical flow estimation subject to physical constraints. Admissible flow fields are restricted to vector fields satifying the Stokes equation. The latter equation includes control variables that allow to control the optical flow so as to fit to the apparent velocities of particles in a given image pair. We show that when the real unknown flow observed through image measurements conforms to the physical assumption underlying the Stokes equation, the control variables allow for a physical interpretation in terms of pressure distribution and forces acting on the fluid. Although this physical interpretation is lost if the assumptions do not hold, our approach still allows for reliably estimating more general and highly non-rigid flows from image pairs and is able to outperform cross-correlation based techniques.
引用
收藏
页码:61 / 78
页数:18
相关论文
共 27 条
  • [11] KAGA A, 1998, P 8 INT S FLOW VIS, P257
  • [12] Kohlberger T, 2003, LECT NOTES COMPUT SC, V2695, P432
  • [13] LANDAU LD, 1991, HYDRODYNAMIK 6 LEHRB
  • [14] *LAVISION, 2005, DAVIS SOFTWARE INTEL
  • [15] Lions J.-L., 1971, OPTIMAL CONTROL SYST
  • [16] MICHAEWICZ Z, 1994, GENETIC ALGORITHMS
  • [17] AN INVESTIGATION OF SMOOTHNESS CONSTRAINTS FOR THE ESTIMATION OF DISPLACEMENT VECTOR-FIELDS FROM IMAGE SEQUENCES
    NAGEL, HH
    ENKELMANN, W
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1986, 8 (05) : 565 - 593
  • [18] Improvements to PIV image analysis by recognizing the velocity gradients
    Nobach, H
    Tropea, C
    [J]. EXPERIMENTS IN FLUIDS, 2005, 39 (03) : 612 - 620
  • [19] NOBACH H, 2005, P 6 INT S PARTICLE I
  • [20] OGAWARA K, 1997, MOVING LEAST SQUARE