.png){kind=icon}
.png){kind=icon}
Return
Pseudo-random illumination microscopy (P-RIM)
.png){kind=icon}
PRE
AI
P
J
M
M
S
A
DOI：10.1088/1361-6420/ae3f29.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Structured illumination microscopy (SIM) is a well-established mean to obtain super-resolved images of fluorescent biological samples, with a two-fold resolution gain. However, it relies on the precise knowledge of the excitation patterns, which is an issue in the presence of uncontrolled pattern distorsions. Random illumination microscopy (RIM) is an alternate solution, built upon a larger number of unknown, randomly generated excitation patterns of controlled statistics, and on a variance-matching process for the super-resolved reconstruction step. In this paper, we propose pseudo-RIM (P-RIM), a novel approach where the sample is illuminated with a limited number of structured patterns as in SIM, combined with a matching process based on second-order statistics as in RIM. We explore mathematical conditions under which P-RIM has a provable super-resolution capacity comparable to SIM and RIM, and we show how to build valid combinations of pseudo-random illuminations. Several pattern sets commonly used in SIM are also compatible with P-RIM, which allows us to compare the behavior of P-RIM with that of SIM on the same simulated dataset, in the presence of pattern distorsions. This empirical test empirically validates that P-RIM variance data processing is nearly insensitive to pattern distorsions, compared to SIM joint least-squares demodulation.
Keywords:
computational imaging
fluorescence microscopy
super-resolution
pseudo-random
Journal
I
IF:
2.1
Papers:
78
Citations:
8.4K
Organization
Cited Papers