Computer with
image digitizer
Video
Video
Monitor
camera
Shore
Floats
er
Riv w
flo
Field
of view
Shore
Figure 5. Typical setup used in an early PTV system for tracking individual
floats on water surface.
recent hardware and software developments, to
lated drifting ice masses, is rather tedious for
the extent that it now is practicable for determin-
determining velocity vectors for numerous ice
ing and mapping fields of ice movement.
pieces, as in a rubble-ice field. Also, it cannot be
Imaged-based techniques for measurement of
used for measuring subsurface patterns of flow
particle and fluid motion have been used for about
and particulate transport.
two decades to determine two-dimensional dis-
PIV has its origin in a technique known as
tributions of flow velocity in laboratory experi-
laser speckle velocimetry (LSV), which was
ments and models. During the past decade, the tech-
developed as a technique for determining strain
niques have evolved along essentially two lines:
fields in deforming solids. Coherent light (i.e.,
particle tracking velocimetry (PTV) and PIV. The
laser light) scattered from a solid surface creates a
two techniques differ in the image-processing prin-
speckle pattern, which changes as the solid
ciple used for determining velocities.
deforms. Whereas particle tracks usually can be
The origin of PTV stems back to various flow-
interpreted from superimposed images by eye,
visualization techniques (Adrian 1991). It requires
changes in speckle patterns are virtually impos-
low concentrations of seeding particles, each par-
sible to unravel that way. With the development
ticle being tracked individually in successive
of the Young's fringe method (a laser beam illu-
images containing short streaks or tracks that, for
minates a double-exposed specklegram to produce
a known period, can be interpreted to give veloci-
a Young's fringe pattern that reveals information
ties. The progenitor versions of PIV were of the
on the displacements of points on a surface), it
PTV type, because of limitations in data-process-
became possible to ascertain the movement and,
ing capacity of computers. Murthy (1991), for
thereby, velocities of a field of points. The next step,
example, describes the use of a rudimentary PTV
taken in the late 1970s, was to use LSV procedures
technique for obtaining distributions of flow
for measuring fluid motion in various flow situa-
velocity in hydraulic models of lock and dam
tions. This step required seeding a flow with light-
facilities. Figure 5 illustrates the typical setup of
scattering particles, fine enough to move exactly
the system used. A video camera and computer
with the fluid. Practical limitations on the amount
with image digitizing software are used to track
of seed particles (tracers) for use in fluid experi-
the movement of single floats, appropriately
ments led to the recognition (Adrian 1984,
ballasted to give either surface or depth-averaged
Pickering and Halliwell 1984) that LSV imaging
velocities. This technique, though still useful for
of fluid flow, strictly speaking, was not LSV; rather,
tracking the trajectory and velocities of a few, iso-
it was a separate mode of image velocimetry that
8
Previous Page
