Design and evaluation of an FFT-based space-time image velocimetry (STIV) for time-averaged velocity measurement

Abstract

Space-Time Image Velocimetry (STIV) is a time-averaged velocity estimation method, which has been used for nonintrusive flow measurements from micro scale to large scale over the past decade. It takes a testing line as the Interrogation Area (IA), and detects the texture orientation of a generated Space-Time Image to determine the 1D velocity. Since STIV is superior to the Large-Scale Particle Image Velocimetry (LSPIV) in spatial resolution and time complexity, it has great potential in real-time monitoring of river flow. However, its practicality is still limited by the deficiencies of performance evaluation and sensitivity analysis due to its complexity and uncertainty. To simplify the detection of texture orientation in spatial domain, an FFT-STIV method was proposed based on Fast Fourier Transform, which detected the orientation of magnitude spectra in frequency domain. Then, a theoretical evaluation was provided, including measuring accuracy, measuring range and computational complexity. Finally, the sensitivities of image size, sampling rate, edge threshold, calibration accuracy, and tracing condition were analyzed based on a Rolling Painting Model (RPM). Results showed that the random errors of FFT-STIV could be controlled within ± 5%, where the systematic error was about ± 1%.