PhaseNet is a Python package for fast optical aberration estimation from bead images using CNNs trained only on synthetic images. Please see our paper for details (preprint here).

PhaseNet requires Python 3.6 or later.

Install via pip install git+https://github.com/mpicbg-csbd/phasenet.git

1. Setup the config file and train the PhaseNet model as shown in the training notebook.
2. Use the trained network to make predictions on acquired bead images from the microscope as shown in the prediction notebook.

1. Check if the training images are in well agreement with the observed PSF using the PSF notebook.
2. Check the generation of wavefront using the wavefront notebook.