Fully Convolutional Approaches for Numerical Approximation of Turbulent Phases in Solar Adaptive Optics
Information on the correlations from solar Shack–Hartmann wavefront sensors is usually used for reconstruction algorithms. However, modern applications of artificial neural networks as adaptive optics reconstruction algorithms allow the use of the full image as an input to the system intended for es...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-07-01
|
Series: | Mathematics |
Subjects: | |
Online Access: | https://www.mdpi.com/2227-7390/9/14/1630 |
id |
doaj-efff32ca76c94302a565cc38bdd5c5fc |
---|---|
record_format |
Article |
spelling |
doaj-efff32ca76c94302a565cc38bdd5c5fc2021-07-23T13:52:25ZengMDPI AGMathematics2227-73902021-07-0191630163010.3390/math9141630Fully Convolutional Approaches for Numerical Approximation of Turbulent Phases in Solar Adaptive OpticsFrancisco García Riesgo0Sergio Luis Suárez Gómez1Enrique Díez Alonso2Carlos González-Gutiérrez3Jesús Daniel Santos4Department of Physics, University of Oviedo, 33007 Oviedo, SpainInstituto Universitario de Ciencias y Tecnologías Espaciales de Asturias (ICTEA), 33004 Oviedo, SpainInstituto Universitario de Ciencias y Tecnologías Espaciales de Asturias (ICTEA), 33004 Oviedo, SpainInstituto Universitario de Ciencias y Tecnologías Espaciales de Asturias (ICTEA), 33004 Oviedo, SpainDepartment of Physics, University of Oviedo, 33007 Oviedo, SpainInformation on the correlations from solar Shack–Hartmann wavefront sensors is usually used for reconstruction algorithms. However, modern applications of artificial neural networks as adaptive optics reconstruction algorithms allow the use of the full image as an input to the system intended for estimating a correction, avoiding approximations and a loss of information, and obtaining numerical values of those correlations. Although studied for night-time adaptive optics, the solar scenario implies more complexity due to the resolution of the solar images potentially taken. Fully convolutional neural networks were the technique chosen in this research to address this problem. In this work, wavefront phase recovery for adaptive optics correction is addressed, comparing networks that use images from the sensor or images from the correlations as inputs. As a result, this research shows improvements in performance for phase recovery with the image-to-phase approach. For recovering the turbulence of high-altitude layers, up to 93% similarity is reached.https://www.mdpi.com/2227-7390/9/14/1630fully convolutional neural networksartificial intelligenceartificial neural networksadaptive opticssolar physics |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Francisco García Riesgo Sergio Luis Suárez Gómez Enrique Díez Alonso Carlos González-Gutiérrez Jesús Daniel Santos |
spellingShingle |
Francisco García Riesgo Sergio Luis Suárez Gómez Enrique Díez Alonso Carlos González-Gutiérrez Jesús Daniel Santos Fully Convolutional Approaches for Numerical Approximation of Turbulent Phases in Solar Adaptive Optics Mathematics fully convolutional neural networks artificial intelligence artificial neural networks adaptive optics solar physics |
author_facet |
Francisco García Riesgo Sergio Luis Suárez Gómez Enrique Díez Alonso Carlos González-Gutiérrez Jesús Daniel Santos |
author_sort |
Francisco García Riesgo |
title |
Fully Convolutional Approaches for Numerical Approximation of Turbulent Phases in Solar Adaptive Optics |
title_short |
Fully Convolutional Approaches for Numerical Approximation of Turbulent Phases in Solar Adaptive Optics |
title_full |
Fully Convolutional Approaches for Numerical Approximation of Turbulent Phases in Solar Adaptive Optics |
title_fullStr |
Fully Convolutional Approaches for Numerical Approximation of Turbulent Phases in Solar Adaptive Optics |
title_full_unstemmed |
Fully Convolutional Approaches for Numerical Approximation of Turbulent Phases in Solar Adaptive Optics |
title_sort |
fully convolutional approaches for numerical approximation of turbulent phases in solar adaptive optics |
publisher |
MDPI AG |
series |
Mathematics |
issn |
2227-7390 |
publishDate |
2021-07-01 |
description |
Information on the correlations from solar Shack–Hartmann wavefront sensors is usually used for reconstruction algorithms. However, modern applications of artificial neural networks as adaptive optics reconstruction algorithms allow the use of the full image as an input to the system intended for estimating a correction, avoiding approximations and a loss of information, and obtaining numerical values of those correlations. Although studied for night-time adaptive optics, the solar scenario implies more complexity due to the resolution of the solar images potentially taken. Fully convolutional neural networks were the technique chosen in this research to address this problem. In this work, wavefront phase recovery for adaptive optics correction is addressed, comparing networks that use images from the sensor or images from the correlations as inputs. As a result, this research shows improvements in performance for phase recovery with the image-to-phase approach. For recovering the turbulence of high-altitude layers, up to 93% similarity is reached. |
topic |
fully convolutional neural networks artificial intelligence artificial neural networks adaptive optics solar physics |
url |
https://www.mdpi.com/2227-7390/9/14/1630 |
work_keys_str_mv |
AT franciscogarciariesgo fullyconvolutionalapproachesfornumericalapproximationofturbulentphasesinsolaradaptiveoptics AT sergioluissuarezgomez fullyconvolutionalapproachesfornumericalapproximationofturbulentphasesinsolaradaptiveoptics AT enriquediezalonso fullyconvolutionalapproachesfornumericalapproximationofturbulentphasesinsolaradaptiveoptics AT carlosgonzalezgutierrez fullyconvolutionalapproachesfornumericalapproximationofturbulentphasesinsolaradaptiveoptics AT jesusdanielsantos fullyconvolutionalapproachesfornumericalapproximationofturbulentphasesinsolaradaptiveoptics |
_version_ |
1721287299124166656 |