Learning from simulation: An end-to-end deep-learning approach for computational ghost imaging

TLDR

AI-based computational imaging typically requires large experimentally collected labeled datasets for training neural networks. The study demonstrates that a practical neural network for computational imaging can be trained solely on simulation data, a strategy applicable to many DL-based imaging solvers. Using computational ghost imaging as a test case, the authors built a one‑step end‑to‑end neural network trained on simulated data to reconstruct 2‑D images directly from 1‑D bucket signals without illumination patterns. This approach is especially advantageous for image transmission through quasi‑static scattering media, where simulating the scattering process for training is straightforward.

Abstract

Artificial intelligence (AI) techniques such as deep learning (DL) for computational imaging usually require to experimentally collect a large set of labeled data to train a neural network. Here we demonstrate that a practically usable neural network for computational imaging can be trained by using simulation data. We take computational ghost imaging (CGI) as an example to demonstrate this method. We develop a one-step end-to-end neural network, trained with simulation data, to reconstruct two-dimensional images directly from experimentally acquired one-dimensional bucket signals, without the need of the sequence of illumination patterns. This is in particular useful for image transmission through quasi-static scattering media as little care is needed to take to simulate the scattering process when generating the training data. We believe that the concept of training using simulation data can be used in various DL-based solvers for general computational imaging.