English

Deep network series for large-scale high-dynamic range imaging

Instrumentation and Methods for Astrophysics 2023-09-28 v3 Machine Learning

Abstract

We propose a new approach for large-scale high-dynamic range computational imaging. Deep Neural Networks (DNNs) trained end-to-end can solve linear inverse imaging problems almost instantaneously. While unfolded architectures provide robustness to measurement setting variations, embedding large-scale measurement operators in DNN architectures is impractical. Alternative Plug-and-Play (PnP) approaches, where the denoising DNNs are blind to the measurement setting, have proven effective to address scalability and high-dynamic range challenges, but rely on highly iterative algorithms. We propose a residual DNN series approach, also interpretable as a learned version of matching pursuit, where the reconstructed image is a sum of residual images progressively increasing the dynamic range, and estimated iteratively by DNNs taking the back-projected data residual of the previous iteration as input. We demonstrate on radio-astronomical imaging simulations that a series of only few terms provides a reconstruction quality competitive with PnP, at a fraction of the cost.

Keywords

Cite

@article{arxiv.2210.16060,
  title  = {Deep network series for large-scale high-dynamic range imaging},
  author = {Amir Aghabiglou and Matthieu Terris and Adrian Jackson and Yves Wiaux},
  journal= {arXiv preprint arXiv:2210.16060},
  year   = {2023}
}

Comments

Accepted for publication in IEEE Proc. ICASSP 2023

R2 v1 2026-06-28T04:42:44.665Z