Image Velocimetry using Direct Displacement Field estimation with Neural Networks for Fluids

TL;DR

This paper introduces a neural network-based method for fluid flow velocity estimation from images that does not require prior training, enabling high-resolution, continuous displacement field predictions validated on synthetic and experimental data.

Contribution

The proposed approach uniquely estimates displacement fields directly from image pairs without prior training, enhancing spatial resolution in fluid velocity measurements.

Findings

Accurate velocity field estimation from synthetic and experimental images.

Effective calculation of turbulence quantities and power spectral density.

No prior training needed, enabling immediate application.

Abstract

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.