Adaptive and automatic diffraction order filtering by singular value decomposition in off-axis digital holographic microscopy

Digital holography is a widely used in many fields for imaging, display and for metrology by exploiting its the capability furnish quantitative phase contrast map. The entire processing pipeline that permits to achieve phase-contrast images can be obtained by a cascade of numerical processing, such as the zero-order and twin image suppression, the automatic refocusing, the phase extraction by aberrations compensation and, if necessary, the phase unwrapping. In this paper, we propose a new method, based on the Singular Value Decomposition (SVD) filtering, to suppress zero-order and twin image in off-axis configuration, thus selecting automatically the desired real diffraction order. Actually, we demonstrate the proposed approach in the case of lack of knowledge about the reference beam's frequency and curvature, which typically occurs in portable off-axis holographic microscope systems for lab-on-a-chip applications. We validate the proposed strategy by a comparison with the common Fourier spatial filtering in the case of different experimental conditions and for several biological samples.