River Monitoring with Ground-based Videography - Fluvial Remote Sensing for Science and Management - page 356

16

River Monitoring with
Ground-based Videography

Bruce J. MacVicar 1 , Alexandre Hauet 2 , Normand Bergeron 3 ,
Laure Tougne 4 and Imtiaz Ali 4, 5
1 Department of Civil and Environmental Engineering, University of Waterloo,
Ontario, Canada
2 EDF - DTG - CHPMC, Toulouse, France
3 Institut National de Recherche Scientifique, Centre Eau Terre et Environnement,
Quebec, Canada
4 Universitey de Lyon, Bron cedex, France
5 Optics Labs, Islamabad, Pakistan

advantage of the inherent advantages of videography for
the monitoring of natural processes in the field.
In rivers, the use of videography for measuring surface
water velocities has been intensively investigated. In
mid 1990s, Japanese teams from the University of Kobe
(Fujita and Komura, 1988, 1994, Aya et al., 1995, Fujita
and Aya, 2000, Fujita et al., 2007a, Fujita et al., 2007b)
first used image velocimetry for velocity measurements in
a riverine environment. Major developments have been
realised by teams at the University of Iowa, USA (Bradley
et al., 2002, Muste et al., 2004, Muste et al., 2005, Muste
et al., 2008) the national engineering research center
(Institut National Polytechnique) in Grenoble, France
(Creutin et al., 2003, Fourquet, 2005, Hauet, 2006, Hauet
et al., 2006, Hauet et al., 2007, Hauet et al., 2008a),
Cemagref, France (Jodeau et al., 2008), and other groups
(e.g. Muller et al., 2002, Harpold and Mostaghimi,
2004). As most of the measurements were taken over
surfaces much larger than those in traditional PIV, the
technique was dubbed Large-Scale PIV (LSPIV) (Fujita
and Hino, 2003). Applications of this approach include

16.1 Introduction

Videography refers to the capture of a series of images.
It is distinguished from the capture of still images by
the fact that the images are taken at a sufficient fre-
quency to allow themovement of objects within the image
frame to be distinguished. The key challenge for scientists
and river managers is to develop computer algorithms
that emulate the mental processes by which we iden-
tify objects and visualise movement so that displacement
vectors can be quantified and other useful information
about the objects in the frame extracted. Particle Image
Velocimetry (PIV) and other image analysis techniques
have been used for almost three decades in laboratory
experiments to measure instantaneous velocity vectors
in laboratory flows (Adrian, 1991, Raffel et al., 1998).
Videography is increasingly implemented to study pro-
cesses in natural environments (Jodeau et al., 2008, Muste
et al., 2008, MacVicar et al., 2009). However, there are a
number of difficulties related to the weather, image angle,
illumination, and contrast that must be overcome to take