When Deep Learning Meets Digital Image Correlation

TL;DR

This paper introduces StrainNet, a CNN designed to measure displacement and strain fields from images, offering a real-time alternative to traditional Digital Image Correlation with competitive accuracy.

Contribution

The paper presents the development of StrainNet, the first CNN specifically trained to perform displacement and strain measurements akin to DIC, demonstrating its effectiveness.

Findings

StrainNet achieves comparable accuracy to traditional DIC methods.

It operates with significantly reduced computation time.

StrainNet is suitable for real-time measurement applications.

Abstract

Convolutional Neural Networks (CNNs) constitute a class of Deep Learning models which have been used in the recent past to resolve many problems in computer vision, in particular optical flow estimation. Measuring displacement and strain fields can be regarded as a particular case of this problem. However, it seems that CNNs have never been used so far to perform such measurements. This work is aimed at implementing a CNN able to retrieve displacement and strain fields from pairs of reference and deformed images of a flat speckled surface, as Digital Image Correlation (DIC) does. This paper explains how a CNN called StrainNet can be developed to reach this goal, and how specific ground truth datasets are elaborated to train this CNN. The main result is that StrainNet successfully performs such measurements, and that it achieves competing results in terms of metrological performance and computing time. The conclusion is that CNNs like StrainNet offer a viable alternative to DIC, especially for real-time applications.