A displacement-shifted vision-based hybrid particle tracking velocimetry (PTV) technique

Abstract

A displacement-shifted approach is introduced to the vision-based particle tracking velocimetry (VB-PTV) technique described in Lei et al. (Exp Fluids 53(5):1251–1268, 2012), using translational and angular displacements. The particle matching algorithm in VB-PTV is based on a proximity matrix, G ij , which favors short distance particle matches over long distance matches. By modifying the formula used in constructing G ij , particles that lie at the expected location of the match are favored. Two displacement-shifted methods are introduced: the first of which relies on particle image velocimetry estimates of particle displacements and the second of which relies on both the expected displacement and direction of the correct match to construct the proximity matrix. These displacement-shifted algorithms improve performance in high gradient (0.3 px/px and above), high displacement flows (upwards of 20 pixels), broadening the range of flows for which VB-PTV can be used. RMS errors in PTV results are reduced by 33 % or more when these displacement-shifted algorithms are made to the VB-PTV algorithm which is used to process Oseen vortex images. Experimental images of shear layer and the wake region of vortex shedding were used to verify the performances of the proposed methods, and the results are in agreement with the synthetic tests.