A Kalman tracker for super-resolution PIV

Abstract

We describe a new super-resolution method of particle-image velocimetry (PIV) based on Kalman filtering and χ²-testing. Performance of the “super-resolution KC” method is evaluated by Monte-Carlo simulation and by applying the method to measurements of flow fields recorded in the form of double-pulse/single-frame and single-pulse/double-frame particle images. When the images have good contrast, and depending on the intensity of the velocity gradients present in the flow, the super-resolution KC method is able to extract valid measurements from 80 to 100% of the available image pairs. In these tests, the vector yield is increased by more than five times compared to standard PIV analysis.