Fluorescence microscopy, which permits observation the structures in living cells, tissues and even small organisms, plays an indispensable role in life science. However, due to the effect of optical diffraction, the spatial resolution of conventional light microscopy is restricted to approximately half the emission wavelength (Fig. 1). To overcome this limitation, several super-resolution fluorescence microscopy techniques based on single-molecule localization have been developed in past several years, for example, stochastic optical reconstruction microscopy (STORM), photoactivated localization microscopy (PALM) and fluorescence PALM (fPALM). These techniques utilize the on-off switching of fluorescent probes to ensure that each active fluorophore is isolated beyond the range of diffraction-limitation, and finally build a high-resolution image from the precise and accurate positions of many single fluorophores.
However, the single-molecule localization techniques require that the active density of fluorophores in each frame must be kept low and the individual fluorophores does not overlap, causing long imaging time to damage live samples. Thus, the low temporal resolutions limit the application of super-resolution microscopy techniques in live-cell imaging field. Although several methods have been developed by fitting with multiple fluorophores at the region of interest to deal with relatively dense fluorescent data, the localization accuracy of fluorophores will decrease dramatically, as the density of emitters becomes high. We focus on how to accurately and quickly obtain the super-resolution results from high single-frame density images.