Library Subscription: Guest
Begell Digital Portal Begell Digital Library eBooks Journals References & Proceedings Research Collections
Journal of Flow Visualization and Image Processing
SJR: 0.161 SNIP: 0.312 CiteScore™: 0.1

ISSN Print: 1065-3090
ISSN Online: 1940-4336

Journal of Flow Visualization and Image Processing

DOI: 10.1615/JFlowVisImageProc.v7.i2.30
19 pages

A CALIBRATION METHOD OF A TWO-CAMERA SYSTEM IN A FACE-TO-FACE CONFIGURATION DESIGNED FOR FAST-FLOW STUDY

L. Riou
Laboratoire Traitement du Signal et Instrumentation, UMR CNRS 5516, 23 rue du Docteur Paul Michelon, 42023 Saint Etienne, France
G. Jacquet
Laboratoire Traitement du Signal et Instrumentation, UMR CNRS 5516, 23 rue du Docteur Paul Michelon, 42023 Saint Etienne, France
J. Fayolle
Laboratoire Traitement du Signal et Instrumentation, UMR CNRS 5516, 23 rue du Docteur Paul Michelon, 42023 Saint Etienne, France
Jacques Jay
Thermal Science Centre of Lyon (CETHIL - UMR CNRS 5008) National Institute of Applied Sciences of Lyon Lyon, France
R. Fouquet
Laboratoire Traitement du Signal et Instrumentation, UMR CNRS 5516, 23 rue du Docteur Paul Michelon, 42023 Saint Etienne, France

ABSTRACT

The study of high-speed flows with particle image velocimetry (PIV) techniques requires charge-coupled device (CCD) cameras with high image frame rate. However, this kind of camera often has limited performance and is expensive. The use of two standard CCD cameras positioned face to face, on each side of the flow, allows the study of very high-speed flow. This system can produce image pairs at a very short delay of less than 200 ns. It is only limited by the precision of the aperture time of the cameras and by the control system. The recorded images are then transformed in order to be seen with the same point of view. This transformation is determined by the earlier calibration of the two cameras. Classic correlation is then performed on a couple of images belonging to the two cameras. The computed displacement is compared with measures obtained by classic PIV on single camera images. The results are very close (a maximum difference of 0.5 pixel).