Signal to noise characterization of an inverse problem-based algorithm for digital inline holography

Abstract : In-line holography is a 3D imaging technique which has been used for many years, especially in experimental fluid mechanics for the 3D localization and sizing of micro-particles from the acquisition of a single 2D image (hologram). This technique is easily usable in an industrial environment thanks to its simple setup. We have recently presented an algorithm of hologram analysis based on an ``inverse problem'' approach. This method find the best model which can explain the hologram and this is realized iteratively by removing at each step the contribution of the detected particle. This method can overcome some limitations of classical approach: like the enlargement of the accessible studied field. Nevertheless, some questions remain on the limitations of the method. We propose in this paper an analysis of the evolution of the signal to noise ratio to clarify the limitations like the size of the studied field, the effect of the cleaning of already detected particles during the process and the influence of the noise generated by the other particles.