Dynamic deformation measurement is a hot topic in optical interferometry research. Currently, most proposed solutions
in this field cannot simultaneously meet the fundamental requirements of accuracy and robustness, because these
methods assume that the speckle field is constant during deformation or utilize iteration algorithm that generates a great
deal of computation. In this paper, an improved shearography technique is presented. The fringe pattern is generated by
the product of the speckle images preprocessed by frequency filter, and then the phase related to the deformation can be
extracted from fringe pattern. Since only one image in deformed stage is used, the proposed method can be well applied
for dynamic deformation measurement. Moreover, the proposed method has much more immunity to the fluctuation of
speckle field compared with conventional method.
In the field of optical measurement, phase always represents the physical quantity to be measured. Thus, phase retrieval from a fringe pattern is a key step for quantitative measurement and evaluation. Much research work has been conducted to develop phase evaluation methods such as fringe tracking and fringe skeletons in earlier, and the more precise methods of phase-shifting and Fourier transform more recently. For phase evaluation, the phase-shifting method requires three or more phase-shifted speckle patterns at each deformed stage; thus, it is not suitable for measurement of continuous deformation. The Fourier transform, on the other hand, requires a high-frequency carrier for phase separation in the spectral domain, which places an additional requirement on experimental arrangement. Thus, it would be desirable to develop a convenient method that can retrieve the modulated phase from a single fringe pattern. We propose an approach that utilizes the phase-clustering property to extract phase information from a single interference specklegram. To explore the ability and limitation for the proposed technique, typical shearographic fringe patterns are used for phase evaluation. Results obtained are similar to those from the standard four-step phase-shifting method. Nonrepeatable continuous movement is also measured by the proposed method, and the results confirm the robustness and accuracy of the clustering method.