Super-resolution and denoising of fluid flow using physics-informed convolutional neural networks without high-resolution labels

TL;DR

This paper introduces a physics-informed convolutional neural network approach for super-resolution and denoising of fluid flow data, capable of producing high-resolution flow fields from low-resolution inputs without requiring high-resolution labels.

Contribution

The study develops a novel CNN-based super-resolution method that leverages physical laws, enabling training without high-resolution labels and applicable to scenarios with partially known physics.

Findings

Effective super-resolution of fluid flows demonstrated on cardiovascular data

Capable of denoising MRI noise types, including Gaussian and non-Gaussian

Unifies forward super-resolution and inverse data assimilation processes

Abstract

High-resolution (HR) information of fluid flows, although preferable, is usually less accessible due to limited computational or experimental resources. In many cases, fluid data are generally sparse, incomplete, and possibly noisy. How to enhance spatial resolution and decrease the noise level of flow data is essential and practically useful. Deep learning (DL) techniques have been demonstrated to be effective for super-resolution (SR) tasks, which, however, primarily rely on sufficient HR labels for training. In this work, we present a novel physics-informed DL-based SR solution using convolutional neural networks (CNN), which is able to produce HR flow fields from low-resolution (LR) inputs in high-dimensional parameter space. By leveraging the conservation laws and boundary conditions of fluid flows, the CNN-SR model is trained without any HR labels. Moreover, the proposed CNN-SR solution unifies the forward SR and inverse data assimilation for the scenarios where the physics is partially known, e.g., unknown boundary conditions. Several flow SR problems relevant to cardiovascular applications have been studied to demonstrate the proposed method's effectiveness and merit. Both Gaussian and non-Gaussian MRI noises are investigated to illustrate the denoising capability.