Image Velocimetry using Direct Displacement Field estimation with Neural Networks for Fluids
This provides an alternative method for experimental fluids mechanics researchers to improve spatial resolution in velocity field estimation, though it appears incremental as it builds on existing PIV methodologies.
The paper tackles the problem of increasing spatial resolution in Particle Image Velocimetry (PIV) by proposing a neural network approach that uses the optical flow equation to estimate displacement vectors between sequential images, resulting in a continuous representation that achieves accurate velocity field estimations and turbulence quantities.
An important tool for experimental fluids mechanics research is Particle Image Velocimetry (PIV). Several robust methodologies have been proposed to perform the estimation of velocity field from the images, however, alternative methods are still needed to increase the spatial resolution of the results. This work presents a novel approach for estimating fluid flow fields using neural networks and the optical flow equation to predict displacement vectors between sequential images. The result is a continuous representation of the displacement, that can be evaluated on the full spatial resolution of the image. The methodology was validated on synthetic and experimental images. Accurate results were obtained in terms of the estimation of instantaneous velocity fields, and of the determined time average turbulence quantities and power spectral density. The methodology proposed differs of previous attempts of using machine learning for this task: it does not require any previous training, and could be directly used in any pair of images.