Noise in laser speckle correlation and imaging techniques

TL;DR

This paper analyzes how noise affects measurements in laser speckle correlation and imaging, revealing that noise scales inversely with the number of pixels and depends on pixel correlations and averaging effects.

Contribution

It provides a quantitative understanding of noise scaling in speckle measurements, guiding improved experimental design and data analysis.

Findings

Noise scales as 1/N with the number of pixels

Noise is sensitive to pixel correlations and averaging methods

Results aid in estimating noise levels in speckle-based techniques

Abstract

We study the noise of the intensity variance and of the intensity correlation and structure functions measured in light scattering from a random medium in the case when these quantities are obtained by averaging over a finite number N of pixels of a digital camera. We show that the noise scales as 1/N in all cases and that it is sensitive to correlations of signals corresponding to adjacent pixels as well as to the effective time averaging (due to the finite sampling time) and spatial averaging (due to the finite pixel size). Our results provide a guide to estimation of noise level in such applications as the multi-speckle dynamic light scattering, time-resolved correlation spectroscopy, speckle visibility spectroscopy, laser speckle imaging etc.