Intrinsic speckle noise in in-line particle holography

Abstract

In-line holography of particle fields suffers from image deterioration caused by intrinsic speckle noise. This is not accounted for by previous theoretical treatments based on the scattering of a single particle, nor has there been any quantitative description of this noise except for an empirical criterion by Royer [ Nouv. Rev. Opt.5, 87 ( 1974)] based on geometrical obscuration. We develop a theoretical model for in-line holography of multiple particles, using diffraction theory and statistical analysis, and show that the virtual image of the particle ensemble is the dominant source of speckle in reconstruction. We quantify the effect of speckle with a signal-to-noise ratio (SNR). The SNR is found to depend on a speckle parameter (which embodies particle diameter, concentration, and sample depth) and on the film gamma. Experimental results show reasonably good agreement with our model. The SNR equation provides prediction of image quality and thence application limits of in-line holography for particle fields. The fundamental understanding obtained here points not only to constraints but also to possible improvements in experimental procedures.