Particle image velocimetry (PIV) and digital image correlation (DIC) are widely used experimental techniques in fluid mechanics and structural dynamics, respectively. PIV is capable of resolving detailed velocity fields around structures, from which the hydrodynamic loading can be reconstructed. However, PIV is ill-suited to capturing the structural response, which is critical for a complete understanding of the bidirectional coupling between the fluid flow and the structural dynamics. On the other hand, DIC can accurately quantify local deformation of the structure, but does not afford the precise identification of the hydrodynamic loading due to the ill-posed nature of the inverse load estimation process. Here, we explore the feasibility of a combined PIV/DIC technique for the investigation of fluid-structure interactions. Specifically, we study fluid-structure interactions associated with a flexible cantilever plate immersed in a steady unidirectional flow. We demonstrate that the combination of pressure estimation from PIV and deformation measurement through DIC enables the precise identification of the hydrodynamic loading and structural response. The proposed methodology may help in improving our understanding of a number of fluid-structure interaction problems, such as biomimetic propulsion, aeroelasticity of airfoils, and hydrodynamic impact on marine structures.

By Peng Zhang

Yes, Particle Image Velocimetry (PIV) and Digital Image Correlation (DIC) are two distinct techniques used in the field of experimental fluid mechanics and materials testing. While they have some similarities, there are also fundamental differences between the two.

Particle Image Velocimetry (PIV):

PIV is a technique used to measure the velocity field of a fluid flow by tracking the movement of particles suspended in the flow. It typically involves illuminating the particles with a laser sheet, capturing sequential images of the particles using a high-speed camera, and then analyzing the images to extract velocity information. PIV is widely used in fluid dynamics research to study phenomena such as flow patterns, turbulence, and boundary layer analysis. It provides quantitative information about the flow field, including velocity vectors and flow structures.

Digital Image Correlation (DIC):

DIC, on the other hand, is a technique used to measure the deformation and strain fields in materials under load. It involves capturing a series of images of a specimen or structure under different loading conditions and then comparing the images to determine the displacement and strain of the material. DIC is commonly used in materials science, mechanical engineering, and structural analysis to study the behavior of materials, including their mechanical properties, stress distribution, and failure mechanisms. It provides information about displacements, strains, and material properties based on image correlation algorithms.

Key Differences:

1. Application: PIV is primarily used to study fluid flow, while DIC is used for materials testing and deformation analysis.

2. Object of Measurement: PIV measures the velocity field of a fluid flow by tracking particles, while DIC measures the displacement and strain fields of solid materials.

3. Imaging: PIV typically requires seeding the flow with particles and using specialized imaging techniques to capture the motion of the particles. DIC relies on imaging the deformation of the surface of a solid material, usually through surface patterns or markers.

4. Analysis: PIV analysis focuses on extracting velocity vectors and flow characteristics from the images. DIC analysis involves comparing images to determine displacements, strains, and material properties.

While there may be some overlap in terms of image processing techniques used in both PIV and DIC, their fundamental objectives and applications make them distinct techniques tailored for different experimental scenarios.

Gopal Sharma Thank you for your answer, but like I mentioned, I don't mean the actual techniques or practical applications involved in both methods. Fundamentally, you are just processing consecutive images and determining the amount of change from image to image.

Whether you are determining velocity or strain is just a matter of whether you're analyzing a liquid or a solid. Ultimately, the vector field produced is mathematically the same in either technique, and I don't see a fundamental difference between the two.

Hi Zherui Martinez-Guo

I am so glad to see your question here!

I spent quite some time at the beginning of my PhD on this topic, mainly arguing with my advisor who kept referring to different processing algorithms which are in fact not intrinsic to any of the two techniques.

The ambiguous term Digital Image Correlation clearly does not help, and we should agree here to mention DIC as the specific experimental technique investigating solid surface deformation.

After the extensive review I did back then, I believe there is no difference between PIV and DIC in essence, and they are simply applied to two different fields of experimental research.

To me, PIV can be considered as DIC on particle speckle patterns and DIC as PIV on solid speckle patterns.

However, I will be happy to be proven wrong here.

The main differences then would be in the specific implementation of PIV and DIC to their own range of applications. For instance, loss of pairs (both in-plane and out-of-plane) is often an issue in PIV applications, which should not be the case for DIC. The algorithm hence needs to be adapted to the physics involved.

Yet, from what I found, both PIV and DIC also use essentially the same approaches for processing the data which can be classified into two main groups: local vs global approaches.

In the end, the two approaches (local/global) are different mainly because of the processing scheme they rely on, and they can often be applied to the same images.

The local approaches are based on correlating small windows (also referred to in both communities as subsets, zones of interest, tiles, image patches, and hand regions) between two frames.

Global approaches consider the entire field of view as a whole and typically rely on a model of the physics involved.

One can find additional information in this DIC (https://link.springer.com/article/10.1007/s11340-012-9603-7) and PIV (https://link.springer.com/article/10.1007/s00348-008-0567-4) references.

On the same note, some fluid dynamics researchers confusingly use “PIV” to only designate the local window-based (cross-correlation) approach.

They hence make the distinction between “PIV” and optical flow since that global approach came later in time.

I believe that they are simply two different approaches of the same technique (PIV) which both rely on images of particles to determine the flow velocity.

However, this is only my opinion, and, in the end, it is the community that settles the common way to refer to these techniques and approaches.

Charles Fort Thank you! Based on what (little) I know, ultimately, the similarities between the two seem to far outweigh their field-specific differences. I would personally prefer DIC to be adopted as an umbrella term for any sort of image pair cross-correlation because the math/physics and practical applications are specific to their respective communities.