upU-Net Approaches for Background Emission Removal in Fluorescence Microscopy

Abstract

The physical process underlying microscopy imaging suffers from several issues: some of them include the blurring effect due to the Point Spread Function, the presence of Gaussian or Poisson noise, or even a mixture of these two types of perturbation. Among them, auto–fluorescence presents other artifacts in the registered image, and such fluorescence may be an important obstacle in correctly recognizing objects and organisms in the image. For example, particle tracking may suffer from the presence of this kind of perturbation. The objective of this work is to employ Deep Learning techniques, in the form of U-Nets like architectures, for background emission removal. Such fluorescence is modeled by Perlin noise, which reveals to be a suitable candidate for simulating such a phenomenon. The proposed architecture succeeds in removing the fluorescence, and at the same time, it acts as a denoiser for both Gaussian and Poisson noise. The performance of this approach is furthermore assessed on actual microscopy images and by employing the restored images for particle recognition.

Neural blind deconvolution with Poisson data

“…• micro: the original image is a phantom of size 128 × 128 described in [46]; the PSF consists in a Gaussian blur with standard deviation equal to √ 5 and peak value 0.032; furthermore, PSNR(g) = 24.51, SSIM(g) = 0.84. • synth001: this is a synthetic simulation of real-world microscopy images; the procedure employed for the generation of such image is explained in [47] and the code is available at https://github.com/AleBenfe/upU-net_Perlin. The PSF used for blurring these images is obtained via the software available at http://bigwww.epfl.ch/algorithms/psfgenerator/ (see [48][49][50] for more technical details).…”

Constrained Plug-and-Play Priors for Image Restoration

“…The challenge of reconstructing a high-quality image x ∈ R n from its degraded measurement b ∈ R n is commonly formulated as a linear inverse problem. Such a problem has to be addressed in several imaging frameworks, such as in medicine [1][2][3][4], microscopy [5][6][7], and astronomy [8][9][10]. Although these are different and maybe distant topics, they share a common linear model [11] for the image acquisition process: namely, b = Ax + η, (1) where A ∈ R n×n is a known blur operator called the Point Spread Function (PSF) [12], and η ∈ R n represents additive random noise with a standard deviation of σ η .…”