An all-electronic digital particle-image velocimetry (PIV) system has been developed for use in measuring three-dimensional velocities at numerous points throughout a plane of interest in a supersonic flow. This system includes two high-resolution charge-coupled-device (CCD) video cameras oriented for stereoscopic imaging of the plane of interest. Two pulsed neodymium: yttrium aluminum garnet (Nd:YAG) lasers and associated optics illuminate the plane of interest with a sheet of light at two slightly different times to obtain double-exposure images of seed particles entrained in the flow. In principle, the velocity vector represented by the double-exposure image for each particle can be obtained by dividing the interexposure displacement vector by the interexposure time.

In the Scheimpflug Condition, the object plane, the image plane, and the median plane through the lens all intersect at a common point. This condition offers advantages for stereoscopic viewing in the present system, as explained in the text.

In PIV and similar systems described in previous articles in NASA Tech Briefs, cameras are aimed perpendicularly to the planes of interest to obtain images indicative of two-dimensional velocities in those planes. Because of the stereoscopy, the images obtained in the present system also contain information on the component of velocity perpendicular to the plane of interest. Of the possible stereoscopic arrangements, the one used in this system involves aiming the lenses of both cameras toward a common point on the plane of interest and tilting the image planes in the cameras to satisfy a condition called the "Scheimpflug condition" (see figure). The advantage of the Scheimpflug condition is that all points of the plane of interest are brought into focus on the image planes, with consequent reduction of the requirement for depth of focus. The Scheimpflug condition entails a minor disadvantage in that it introduces some distortion; for example, a suitably oriented rectangle in the object plane becomes imaged to an isosceles trapezoid. Fortunately, a correction for this distortion can be readily incorporated into the image-data-processing algorithm.

The stereoscopic double-exposure images are digitized, the images are divided into regions, and the image data are processed by use of an autocorrelation technique to obtain a candidate-velocity-vector map of the plane of interest. Typically, this map contains a few erroneous vectors. The most probable candidate velocity vectors are selected in a fuzzy inference operation, in a manner similar to that described in "Digital Particle-Image Velocimetry Enhanced by Fuzzy Logic" (LEW-16415), NASA Tech Briefs, Vol. 21, No. 12 (December 1997), page 81. In this operation, the velocity vectors of the five highest correlation peaks (excluding the zero-order peak) in each region are compared with those of the five highest correlation peaks in each of the four surrounding regions. For each region, the velocity vector most similar to the velocity vectors of the selected correlation peaks of the other regions is selected. The justification for selecting velocity vectors on the basis of similarity to adjacent velocity vectors lies in the fundamental continuity of flow.

This work was done by Mark P. Wernet of Lewis Research Center. For further information, access the Technical Support Package (TSP) free on-line at under the Physical Sciences category, or circle no. 129on the TSP Order Card in this issue to receive a copy by mail ($5 charge). Inquiries concerning rights for the commercial use of this invention should be addressed to

NASA Lewis Research Center, Commercial Technology Office, Attn: Tech Brief Patent Status, Mail Stop 7-3, 21000 Brookpark Road, Cleveland, Ohio 44135.

Refer to LEW-16500.