An EM Approach for Time-Variant Poisson-Gaussian Model Parameter Estimation

Abstract

International audienceThe problem of estimating the parameters of a Poisson-Gaussian model from experimental data has recently raised much interest in various applications, for instance in confocal fluorescence microscopy. In this context, a field of independent random variables is observed, which is varying both in time and space. Each variable is a sum of two components, one following a Poisson and the other a Gaussian distribution. In this paper, a general formulation is considered where the associated Poisson process is nonstationary in space and also exhibits an exponential decay in time, whereas the Gaussian component corresponds to a stationary white noise with arbitrary mean. To solve the considered parametric estimation problem, we follow an iterative Expectation-Maximization (EM) approach. The parameter update equations involve deriving finite approximation of infinite sums. Expressions for the maximum error incurred in the process are also given. Since the problem is non-convex, we pay attention to the EM initialization, using a moment-based method where recent optimization tools come into play. We carry out a performance analysis by computing the Cramer-Rao bounds on the estimated variables. The practical performance of the proposed estimation procedure is illustrated on both synthetic data and real fluorescence macroscopy image sequences. The algorithm is shown to provide reliable estimates of the mean/variance of the Gaussian noise and of the scale parameter of the Poisson component, as well as of its exponential decay rate. In particular, the mean estimate of the Poisson component can be interpreted as a good-quality denoised version of the data

Measuring the Noise of Digital Imaging Sensors by Stacking Raw Images Affected by Vibrations and Illumination Flickering

“…A comparison with multi-image estimation techniques such as [20] or [27] would require a separate study. Section 4.1 details a running example.…”

“…A large flickering amplitude ensures a higher estimation accuracy, giving results within the range of the photon transfer method. Such an approach may be useful for multi-image denoising methods (as in [9,34]) or for noise parameter estimation in fluorescence imaging affected by photobleaching (as in [27]), where a nonconstant illumination has been noticed. Table 5 Estimating camera parameters (red channel).…”

A Guided Tour of Selected Image Processing and Analysis Methods for Fluorescence and Electron Microscopy

“…In fluorescence imaging, this situation is not usual but is considered in Time-Correlated Single Photon Counting (TCSPC)-Fluorescence Lifetime Imaging Microscopy (FLIM) [148], [149]. A alternate approach is to consider Poisson noise statistics (or Poisson-Gaussian noise statistics [150]- [152]) and maximum likelihood estimators [153] or Maximum a Posteriori estimators [154]. The idea is to directly handle Poisson noise without "Gaussianization" of the data, which is more appropriate for low SNRs.…”

Noise Parameter Mismatch in Variance Stabilization, With an Application to Poisson–Gaussian Noise Estimation

“…This can be thought as fitting a curve to a scatter plot in the Cartesian plane where the abscissa and ordinate are, respectively, the mean and standard deviation (or variance) of the image samples. In case of a time-variant model (as due to decay of markers in fluorescence imaging), the problem has also been addressed through an expectationmaximization algorithm [9]. Regardless of the particular noise model, the problem of estimating the parameter(s) associated with it can be approached either by taking advantage of multiple images (see [10]), or by considering only a single image.…”