Compressed sensing for digital holographic microscopy

Abstract

This paper describes an original microscopy imaging framework successfully employing Compressed Sensing for digital holography. Our approach combines a sparsity minimization algorithm to reconstruct the image and digital holography to perform quadrature-resolved random measurements of an optical field in a diffraction plane. Compressed Sensing is a recent theory establishing that near-exact recovery of an unknown sparse signal is possible from a small number of non-structured measurements. We demonstrate with practical experiments on holographic microscopy images of cerebral blood flow that our CS approach enables optimal reconstruction from a very limited number of measurements while being robust to high noise levels.