Iterative image deformation methods in PIV

Abstract

Image processing methods for particle image velocimetry (PIV) are reviewed. The discussion focuses on iterative methods aimed at enhancing the precision and spatial resolution of numerical interrogation schemes. Emphasis is placed on the efforts made to overcome the limitations of the correlation interrogation method with respect to typical problems such as in-plane loss of pairs, velocity gradient compensation and correlation peak locking. The discussion shows how the correlation signal benefits from simple operations such as the window-offset technique, or the continuous window deformation, which compensates for the in-plane velocity gradient. The image interrogation process is presented within the discussion of the image matching problem and several algorithms and implementations currently in use are classified depending on the choice made about the particle image pattern matching scheme. Several methods that differ in their implementations are found to be substantially similar. Iterative image deformation methods that account for the continuous particle image pattern transformation are analysed and the effect of crucial choices such as image interpolation method, displacement prediction correction and correlation peak fit scheme are discussed. The quantitative performance assessment made through synthetic PIV images yields order-of-magnitude improvement on the precision of the particle image displacement at a sub-pixel level when the image deformation is applied. Moreover, the issue of spatial resolution is addressed and the limiting factors of the specific interrogation methods are discussed. Finally, an attempt for a flow-adaptive spatial resolution method is proposed. The method takes into account the local velocity derivatives in order to perform a local increase of the interrogation spatial density and a refinement of the local interrogation window size. The resulting spatial resolution is selectively enhanced. The method's performance is analysed and compared with some precursor techniques, namely the conventional cross-correlation analysis with and without the effect of a window discrete offset and deformation. The suitability of the method for the measurement in turbulent flows is illustrated with the application to a turbulent backward facing step flow.

Three-dimensional organisation of primary and secondary crossflow instability

“…At this region, the inherent downwelling action of the primary CF instability results in elevated shear near the wall, as showed in figure 13b. Such high shears are beyond the measurable threshold of 0.5px/px reported by Scarano (2002) for PIV measurements. It is believed that the pronounced errors are not deemed disruptive to the conclusions of this study because they appear in a region not associated with a particular instability mode.…”

“…The large displacement ensures a relatively high dynamic range of the velocity measurement (Adrian & Westerweel 2011), which is suited to enable the measurement of the weak velocity components in planes orthogonal to the streamwise direction. However, it compromises the measurement of the highly sheared regions close to the wall due to the distortion of the particle image pattern beyond 0.5pixels/pixel (Scarano 2002), as it will be discussed in later sections. A dual layer target was used for the calibration of the tomographic imaging system.…”

On the resolution limit of digital particle image velocimetry

“…However, although it is possible to compute velocity vectors even in smaller scales than a micrometer grid, these vectors are not independent and bias errors can occur in particular situations. Even though many different approaches to increase the accuracy and resolution of DPIV were presented, see Adrian and Westerweel (2010), Keane et al (1995), Raffel et al (2007), Scarano (2001), Stanislas et al (2003Stanislas et al ( , 2005Stanislas et al ( , 2008, Stitou and Riethmuller (2001), Willert (1997), for instance, a detailed analysis of the resolution limit is still lacking and will be the focus of this paper.…”

Experimentally investigation of the CCD binning option in particle imaging velocimetry measurements of laminar flows

“…The relative position of particles within the same FOV at a minimum of two instances of time is recorded by employing a high-speed camera. Cross correlation analysis of consecutive frames is performed to measure the average particle displacements within interrogation spots, which Correspondence: basak.akselli@ume.tubitak.gov.tr are then interpolated to ascertain the overall motion, field of the fluid [8][9][10].…”