Deep neural networks in single-shot ptychography.

We develop and explore a deep learning based single-shot ptychography reconstruction method. We show that a deep neural network, trained using only experimental data and without any model of the system, leads to reconstructions of natural real-valued images with higher spatial resolution and better resistance to systematic noise than common iterative algorithms. (C) 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement