FMD stands for Fluorescence Microscopy Denoising, which is a dataset dedicated to Poisson-Gaussian denoising. The dataset consists of 12,000 real fluorescence microscope images. They were obtained from representative biological samples (cells, zebrafish, and mouse brain tissue) using commercial confocal microscopes, two-photon microscopes, and wide-field microscopes. The dataset uses image averaging technology to effectively obtain 60,000 noisy images and benchmark truth images with different noise levels.

This dataset can be used to test denoising algorithms and is an important tool for high-quality, real-time denoising applications in biomedical research.