The potential of quantum computers for Particle Image Velocimetry

Research Question and Approach

Particle Image Velocimetry (PIV) is a standard technique for measuring fluid velocity fields by calculating the cross-correlation between successive images of tracer particles. Classically, this is computationally expensive because it requires millions of Fast Fourier Transforms (FFTs). The authors propose a quantum-based alternative, QuPIV, designed to handle this task more efficiently by leveraging quantum parallelism.

The QuPIV pipeline consists of three main stages: a specialized sparse state preparation, a two-dimensional quantum cross-correlation evaluation, and a modified amplitude amplification process. To address the "quantum von-Neumann bottleneck"—the difficulty of loading classical data into quantum states and extracting results—the authors implement a data reduction strategy that identifies only the most significant particle positions. They further optimize the circuit by using a contracted ground-state projector, which reduces the total number of gates required for amplitude amplification.

The study demonstrates that the QuPIV algorithm can successfully compute velocity fields by identifying the correlation peak, which corresponds to the highest measurement probability. By focusing on the peak position rather than the entire correlation map, the algorithm avoids the need to extract exponentially large amounts of data. Numerical simulations on both synthetic and experimental PIV data confirm that the algorithm can achieve sub-pixel accuracy comparable to classical methods while providing a framework for handling industrial-scale fluid dynamics problems on quantum computers.

As fluid dynamics research moves toward higher spatial and temporal resolutions, the computational cost of classical PIV becomes a significant barrier. QuPIV offers a potential path toward accelerating these calculations. Furthermore, the authors' refinement of amplitude amplification provides a generalizable technique that could improve the efficiency of other quantum algorithms requiring similar probabilistic state manipulation.