Optimal transport for super resolution applied to astronomy imaging
2022 (English)In: 2022 30th European Signal Processing Conference (EUSIPCO), IEEE, 2022, Vol. 272, p. 1971-1975Conference paper, Published paper (Refereed)
Abstract [en]
Super resolution is an essential tool in optics, especially on interstellar scales, due to physical laws restricting possible imaging resolution. We propose using optimal transport and entropy for super resolution applications. We prove that the reconstruction is accurate when sparsity is known and noise or distortion is small enough. We prove that the optimizer is stable and robust to noise and perturbations. We compare this method to a state of the art convolutional neural network and get similar results for much less computational cost and greater methodological flexibility.
Place, publisher, year, edition, pages
IEEE, 2022. Vol. 272, p. 1971-1975
Series
European Signal Processing Conference, ISSN 2219-5491, E-ISSN 2076-1465
Keywords [en]
optimal transport, Wasserstein distance, super resolution, compressed sensing, sparse imaging, sparse regularization, sparsity, maximum entropy, convolutional neural network
National Category
Computational Mathematics
Identifiers
URN: urn:nbn:se:umu:diva-229605DOI: 10.23919/eusipco55093.2022.9909820ISBN: 978-1-6654-6799-5 (print)ISBN: 978-90-827970-9-1 (electronic)ISBN: 978-90-827970-8-4 (electronic)OAI: oai:DiVA.org:umu-229605DiVA, id: diva2:1897802
Conference
30th European Signal Processing Conference (EUSIPCO), Belgrade, Serbia, August 29 - September 2, 2022
Funder
Knut and Alice Wallenberg Foundation, 2018–03572024-09-162024-09-162024-09-16Bibliographically approved