Search

Adaptive optics (AO) is a powerful tool for correcting wavefront errors induced by complex structures of biological samples which significantly causes image degradation. A scene-based sensing technique is being popular in microscopic AO systems with Shack-Hartmann (SH) wavefront sensors. A problem in application of the technique is that the shapes of images observed on SAs vary dependently to their positions on the aperture, especially when using microscopic objectives with higher NAs. To mitigate this problem, a differential sensing technique is used that enables measuring image shifts with high correlations over the aperture. Experiments using an artificial testing target including fluorescent beads, which simulates the leaf of moss, were conducted to investigate imaging performances of the present AO system. Unbiased maximum ratios were measured from blurred and AO-corrected images, and then the Strehl ratios were derived from them. Resultant Strehl ratios were around 0.58.

We develop a machine learning (ML) software to estimate morphological parameters (e.g., the half-light radius re) of high redshift galaxies in the Subaru/Hyper Suprime-Cam data. To make the ML software capture simultaneously galaxy morphological features and point spread function (PSF) broadening effects, we implement a two-stream convolutional neural network (CNN) for inputs of galaxy and PSF images. Thanks to large training samples of galaxy and PSF images, the two-stream CNN estimates re more accurately than a single-stream CNN with only galaxy images. Our ML software would be a useful tool to investigate galaxy morphological properties with PSF-unstable images obtained in future large-area ground-based surveys.