Improved training for the deep learning wavefront sensor

Camilo Weinberger, Felipe Guzmán, Esteban Vera

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations


We have recently proposed the deep learning wavefront sensor, capable of directly estimating Zernike coefficients of aberrated wavefronts from a single intensity image by using a convolutional neural network. However, deep neural networks demand an intensive training stage, where more training examples allow to improve the accuracy and increase the amount of the estimated Zernike modes. Since low order aberrations such as tip and tilt only produce space-invariant motion of the PSF, we propose to treat tip and tilt estimation separately when training the deep learning wavefront sensor, decreasing the training efforts while keeping the wavefront sensor performance. In this paper, we also introduce and test simpler architectures for deep learning wavefront sensing, while exploring the impact of reducing the number of pixels to estimate a given amount of Zernike coefficients. Our preliminary results indicate that we can achieve a significant prediction speedup aiming for real time adaptive optics systems.

Original languageEnglish
Title of host publicationAdaptive Optics Systems VII
EditorsLaura Schreiber, Dirk Schmidt, Elise Vernet
ISBN (Electronic)9781510636835
StatePublished - 2020
EventAdaptive Optics Systems VII 2020 - Virtual, Online, United States
Duration: 14 Dec 202022 Dec 2020

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X


ConferenceAdaptive Optics Systems VII 2020
Country/TerritoryUnited States
CityVirtual, Online


  • Adaptive optics
  • Machine-learning
  • Wavefront sensor


Dive into the research topics of 'Improved training for the deep learning wavefront sensor'. Together they form a unique fingerprint.

Cite this