Article (Scientific journals)
Focal plane wavefront sensing using machine learning: performance of convolutional neural networks compared to fundamental limits
Orban De Xivry, Gilles; Quesnel, Maxime; Vanberg, P.-O. et al.
2021In Monthly Notices of the Royal Astronomical Society, 505 (4), p. 5702-5713
Peer Reviewed verified by ORBi
 

Files


Full Text
OrbandeXivry+21_mnras.pdf
Publisher postprint (1.9 MB)
Download

This article has been accepted for publication in MNRAS ©: 2021 The Authors. Published by Oxford University Press on behalf of he Royal Astronomical Society. All rights reserved.


All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
instrumentation: high angular resolution; adaptive optics; methods: numerical; Astrophysics - Instrumentation and Methods for Astrophysics; Physics - Optics
Abstract :
[en] Focal plane wavefront sensing (FPWFS) is appealing for several reasons. Notably, it offers high sensitivity and does not suffer from non-common path aberrations (NCPAs). The price to pay is a high computational burden and the need for diversity to lift any phase ambiguity. If those limitations can be overcome, FPWFS is a great solution for NCPA measurement, a key limitation for high-contrast imaging, and could be used as adaptive optics wavefront sensor. Here, we propose to use deep convolutional neural networks (CNNs) to measure NCPAs based on focal plane images. Two CNN architectures are considered: ResNet-50 and U-Net that are used, respectively, to estimate Zernike coefficients or directly the phase. The models are trained on labelled data sets and evaluated at various flux levels and for two spatial frequency contents (20 and 100 Zernike modes). In these idealized simulations, we demonstrate that the CNN-based models reach the photon noise limit in a large range of conditions. We show, for example, that the root mean squared wavefront error can be reduced to <λ/1500 for 2 × 106 photons in one iteration when estimating 20 Zernike modes. We also show that CNN-based models are sufficiently robust to varying signal-to-noise ratio, under the presence of higher order aberrations, and under different amplitudes of aberrations. Additionally, they display similar to superior performance compared to iterative phase retrieval algorithms. CNNs therefore represent a compelling way to implement FPWFS, which can leverage the high sensitivity of FPWFS over a broad range of conditions.
Research center :
STAR - Space sciences, Technologies and Astrophysics Research - ULiège
Montefiore Institute - Montefiore Institute of Electrical Engineering and Computer Science - ULiège
Disciplines :
Space science, astronomy & astrophysics
Author, co-author :
Orban De Xivry, Gilles  ;  Université de Liège - ULiège > Département d'astrophys., géophysique et océanographie (AGO) > PSILab
Quesnel, Maxime ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Big Data
Vanberg, P.-O.
Absil, Olivier  ;  Université de Liège - ULiège > Département d'astrophys., géophysique et océanographie (AGO) > PSILab
Louppe, Gilles  ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Big Data
Language :
English
Title :
Focal plane wavefront sensing using machine learning: performance of convolutional neural networks compared to fundamental limits
Publication date :
09 June 2021
Journal title :
Monthly Notices of the Royal Astronomical Society
ISSN :
0035-8711
eISSN :
1365-2966
Publisher :
Oxford University Press, Oxford, United Kingdom
Volume :
505
Issue :
4
Pages :
5702-5713
Peer reviewed :
Peer Reviewed verified by ORBi
European Projects :
H2020 - 819155 - EPIC - Earth-like Planet Imaging with Cognitive computing
Name of the research project :
EPIC; NNExI
Funders :
CER - Conseil Européen de la Recherche [BE]
FWB - Fédération Wallonie-Bruxelles [BE]
CE - Commission Européenne [BE]
Available on ORBi :
since 10 August 2021

Statistics


Number of views
97 (15 by ULiège)
Number of downloads
66 (6 by ULiège)

Scopus citations®
 
14
Scopus citations®
without self-citations
11
OpenCitations
 
3

Bibliography


Similar publications



Contact ORBi