Three-Dimensional Particle Tracking Velocimetry Based on Automated Digital Image Processing

Abstract

A three-dimensional particle tracking velocimeter for measuring an instantaneous distribution of all the three velocity components in a liquid flow is developed using an automated digital image processing technique. The measurement system consists of three TV cameras, a digital image processor, a laser disk recorder and a 16-bit microcomputer. Motions of neutrally buoyant tracer particles introduced into the flow field are observed by the cameras, and the three-dimensional displacement of each particle is calculated from consecutive TV frames recorded on the laser disk. Instantaneous velocity profiles in an unsteady laminar Couette flow between two concentric cylinders are measured. The results are in good agreement with predictions within the measurement uncertainties evaluated systematically. Furthermore, the measurement of a decaying turbulence in a stirred water tank demonstrates that the present technique is applicable to turbulent flows.

A Neural Network Based Algorithm for Particle Pairing Problem of PIV Measurements

“…The 3-D coordinates of particles used for the matching operation are obtained from the numerical database of the standard images. But, it should be noted that, in the fully 3-D system the accuracy of the matching is dependant upon the accuracy of the algorithms to extract the three dimensional coordinates of the particles [32]- [34]. But in any case, the constraint conditions proposed here are, obviously, more favourable to three-dimensional flows as particle positions in this case are resolved in depth directions also and hence provide the more accurate estimates of distance parameters.…”

Stereoscopic Correspondence of Particles for 3-Dimensional Particle Tracking Velocimetry by using Genetic Algorithm

“…This results using conventional method (particle pairing without GA) come from particle matching using a conventional epipolar-line proximity analysis based on Nishino et al [10] where the particles i and k which are able to produce the minimum value of d ik + d ki are considered as perfect pairs. Table 2, it is recognized that the correct pair rate is more or less increased with the use of the genetic algorithm, regardless of which series of image datasets and frames are used.…”

“…This leads to a low density of measurement velocity vectors that looks much less satisfactory than those of the 2-D particle tracking velocimetry. In this context, for full-volume 3-D flow measurement, the epipolar line nearest-neighbor analysis [10] seems a standard technique of 3-D particle pairing, although there are some variations or modifications (simplified or complicated) reported so far in the literature. However, even within the extent of the standard epipolar line nearest-neighbor analysis, the detail of the algorithm is different from author to author.…”

Image processing for the study of bedload transport of two-size spherical particles in a supercritical flow

“…Depending on the application and on the assumptions made on point displacement, more or less complex approaches have been proposed (Sethi and Jain 1987;Hwang 1989;Salari and Sethi 1990). A particular application known as PTV is to determine the velocity field of a fluid carrying small particles (Nishino et al 1989;Economikos et al 1990;Fayolle et al 1996;Udrea et al 2000). In our study, we focused on the Fig.…”