Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Image-based wavefront sensing for astronomy using neural networks

Andersen, Torben LU ; Owner-Petersen, Mette LU and Enmark, Anita LU (2020) In Journal of Astronomical Telescopes, Instruments, and Systems 6(3).
Abstract
Motivated by the potential of nondiffraction limited, real-time computational image sharpening with neural networks in astronomical telescopes, we studied wavefront sensing with convolutional neural networks based on a pair of in-focus and out-of-focus point spread functions. By simulation, we generated a large dataset for training and validation of neural networks and trained several networks to estimate Zernike polynomial approximations for the incoming wavefront. We included the effect of noise, guide star magnitude, blurring by wide-band imaging, and bit depth. We conclude that the “ResNet” works well for our purpose, with a wavefront RMS error of 130 nm for r0  =  0.3  m, guide star magnitudes 4 to 8, and inference time of 8 ms. It... (More)
Motivated by the potential of nondiffraction limited, real-time computational image sharpening with neural networks in astronomical telescopes, we studied wavefront sensing with convolutional neural networks based on a pair of in-focus and out-of-focus point spread functions. By simulation, we generated a large dataset for training and validation of neural networks and trained several networks to estimate Zernike polynomial approximations for the incoming wavefront. We included the effect of noise, guide star magnitude, blurring by wide-band imaging, and bit depth. We conclude that the “ResNet” works well for our purpose, with a wavefront RMS error of 130 nm for r0  =  0.3  m, guide star magnitudes 4 to 8, and inference time of 8 ms. It can also be applied for closed-loop operation in an adaptive optics system. We also studied the possible use of a Kalman filter or a recurrent neural network and found that they were not beneficial to the performance of our wavefront sensor. (Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Neural networks, Point spread functions, Wavefront sensors, Wavefronts, Stars, Astronomy, Telescopes, V-band, Error analysis
in
Journal of Astronomical Telescopes, Instruments, and Systems
volume
6
issue
3
article number
034002
pages
15 pages
publisher
SPIE
external identifiers
  • scopus:85090380932
ISSN
2329-4124
DOI
10.1117/1.JATIS.6.3.034002
language
English
LU publication?
yes
id
96b1b336-f865-4325-9b4c-7f53471adb65
date added to LUP
2020-07-13 23:27:42
date last changed
2024-04-03 10:02:26
@article{96b1b336-f865-4325-9b4c-7f53471adb65,
  abstract     = {{Motivated by the potential of nondiffraction limited, real-time computational image sharpening with neural networks in astronomical telescopes, we studied wavefront sensing with convolutional neural networks based on a pair of in-focus and out-of-focus point spread functions. By simulation, we generated a large dataset for training and validation of neural networks and trained several networks to estimate Zernike polynomial approximations for the incoming wavefront. We included the effect of noise, guide star magnitude, blurring by wide-band imaging, and bit depth. We conclude that the “ResNet” works well for our purpose, with a wavefront RMS error of 130 nm for r0  =  0.3  m, guide star magnitudes 4 to 8, and inference time of 8 ms. It can also be applied for closed-loop operation in an adaptive optics system. We also studied the possible use of a Kalman filter or a recurrent neural network and found that they were not beneficial to the performance of our wavefront sensor.}},
  author       = {{Andersen, Torben and Owner-Petersen, Mette and Enmark, Anita}},
  issn         = {{2329-4124}},
  keywords     = {{Neural networks; Point spread functions; Wavefront sensors; Wavefronts; Stars; Astronomy; Telescopes; V-band; Error analysis}},
  language     = {{eng}},
  month        = {{07}},
  number       = {{3}},
  publisher    = {{SPIE}},
  series       = {{Journal of Astronomical Telescopes, Instruments, and Systems}},
  title        = {{Image-based wavefront sensing for astronomy using neural networks}},
  url          = {{http://dx.doi.org/10.1117/1.JATIS.6.3.034002}},
  doi          = {{10.1117/1.JATIS.6.3.034002}},
  volume       = {{6}},
  year         = {{2020}},
}