Phase analysis of interference signal with optical Hilbert transform based on orthogonal linear polarization phase shifting

Digital optical interferometry is a whole-field technique for deformation measurements with high temporal and spatial resolution and with possibility for in-situ measurements of dynamic events. Quantitative evaluation of the deformation field is obtained by analysing the phase of the interference signal. To accurately calculate the phase, at least three phase-shifted interference patterns are required. Consequently, majority of the phase-analyzing algorithms that were proposed in the past were applicable only for static events [1]. For dynamic events, there is a different approach, in which each point in the image is considered independently, and the interference signal in time domain is analyzed [2, 3]. In our previous studies, we proposed a Hilbert transform (HT) method for phase analyses [3]. The method has the advantage of low noise level in the signal, in a fully automated manner. One disadvantage of HT method, however, is the accumulation of huge three-dimensional data sets that requires considerable amount of computer memory. In addition, the processing is carried out after the experiment, which does not allow in-situ measurements of deformation fields. To overcome these problems, we proposed a method that uses both temporal and spatial interference signal, where HT is carried out optically by spatial phase-shifting, which is achieved by separation of orthogonal components of polarized light. Similar methods have also been proposed in the past [4-7]. In the present method, only one measurement is necessary to calculate the phase distribution. For dynamic events, the temporal information of the interference signal can used to determine the bias intensity and to unwrap the phase in time domain.