Abstract
Of all the two-beam interference patterns, the ones obtained in speckle interferometry (SI) are the most difficult to be phase-demodulated. Many solutions exist in classical smooth-wave interferometry and alike techniques, both in static and dynamic regimes. In SI, the three constituents of the signals - the background, the modulation and the phase - are all basically random variables. There is no way to make a prediction of the evolution of these variables outside the small size of the correlation volumes - the volumes defined by the average speckle grain. To some extent, the classical methods can be adapted to SI. Here, we prefer to develop a series of new processing tools tailored to the specificities of the dynamic SI signals: the cooperative use of the empirical mode decomposition (EMD), the Hilbert transform (HT), and the three dimensional piecewise processing (3DPP) for recovering efficiently the phase of these signals.
Highlights
During the elapsed decade, photomechanics has established itself as a prominent discipline of experimental mechanics
Considering that dynamic regimes are the more frequent and that processing schemes in this case are the most open to improvements, we focused our developments on the processing of long sequences of recorded specklegrams, obtained in continuous non-periodic loading regimes
2.4 Proposed road map for extracting the phase of dynamic speckle interferometry (SI) signals In SI, two broad categories of signals should be distinguished, due to their peculiar appearance and features: the signals given by each pixel along the temporal axis, and those resulting from any 1D spatial cut made in the correlation fringes, i.e. the pattern obtained by the absolute subtraction, or similar arithmetic operations, of two correlated specklegrams associated to two states of the object deformation
Summary
Photomechanics has established itself as a prominent discipline of experimental mechanics. DIC stands for “digital image correlation” and SI for “speckle interferometry” The impact of both types of methods strongly depends on the ability to process the rough acquired data, i.e. to convert intensity images and fringe patterns into sets of mechanically interesting variables. This article proposes a complete solution of this problem in the case of SI applied to the analysis of the deformation of solid bodies subjected to dynamic loadings. The last section presents a complete example of the joint utilization of these three processing tools and summarizes their main features
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
