Simulations of Single Molecule Localization Microscopy frames with scattered single emitters

Datasets used in the work "Combining deep learning with SUPPOSe and Compressed Sensing for SNR-enhanced localization of overlapping emitters".

The file Sample dataset.zip contains simulated images of frames of Single-Molecule Localization Microscopy. In the directory structure, Q is the number of emitters, d is the emitter distance (in pixels), imax is the image maximum intensity and i indexes different noise realizations. There are three images within each folder: X is a noiseless image, Y is the image with noise and background and Z is a denoised image predicted using a convolutional neural network.

The file Train dataset.zip contains 5000 simulated pairs of images of single emitters distributed randomly that were used to train a convolutional neural network for denoising. The folder X contains noiseless images and the folder Y contains the corresponding images with noise and background.

In all cases, a Gaussian PSF with size \(\sigma = 3\) px was used as the PSF of the imaging system, Noise is modeled as a Poisson process with a dark signal \(i_{dark} = 10\).

Corresponding author: Axel M. Lacapmesure ([e-mail])

CHANGELIST

Version 2: corrected all file extensions in "Train dataset.zip" that were wrong.