Single image sensorless adaptive optics for microscopy
Biwei Zhang, Qi Hu, Jingyu Wang, Martin J. Booth
公開日: 2025/9/22
Abstract
Images obtained by microscopes are generally degraded by aberrations. Adaptive optics (AO) has been widely used in to compensate for aberrations and improve reduced image quality. Requiring no separate wavefront sensor, sensorless AO methods deduce the aberrations from a sequence of images acquired with different phase modulations. Despite that these methods are versatile and flexible, they involve sequential acquisition of multiple images thus time-consuming to implement. Here, we propose a new sensorless AO method based on only a single image. Experimental results on a two-photon fluorescence microscope showed that our method achieved distinct correction while imaging different types of structures under various noise levels with each correction loop taking less than 1 second. This work has significantly accelerated sensorless AO, which will assist its application in many challenging imaging scenarios of microscopy.