Phase retrieval in digital speckle pattern interferometry using a chirped gaussian wavelet transform, and a smoothed time-frequency distribution

In this paper we analyze the application of a chirped Gaussian wavelet transform to retrieve the phase distribution in digital speckle pattern interferometry (DSPI). The performance of this method is evaluated using computer-simulated fringes, approach that allows an accurate determination of the phase map encoded by the fringe pattern. It is shown that the local chirp-based approximation does not give good results in the neighborhood of stationary phase points. To overcome this limitation, we compare the wavelet analysis by applying a smoothed time-frequency distribution. We show that the application of this distribution allows the determination of the phase map where the local chirp-based approximation fails. We also discuss the influence of the filtering process to smooth the DSPI fringes and some additional limitations that emerge when these methods are applied.