Learning to do multiframe wavefront sensing unsupervisedly: applications to blind deconvolution

Abstract

Observations from ground based telescopes are affected by the presence of the Earth atmosphere, which severely perturbs them. The use of adaptive optics techniques has allowed us to partly beat this limitation. However, image selection or post-facto image reconstruction methods applied to bursts of short-exposure images are routinely needed to reach the diffraction limit. Deep learning has been recently proposed as an efficient way to accelerate these image reconstructions. Currently, these deep neural networks are trained with supervision, so that either standard deconvolution algorithms need to be applied a-priori or complex simulations of the solar magneto-convection need to be carried out to generate the training sets. Our aim here is to propose a general unsupervised training scheme that allows multiframe blind deconvolution deep learning systems to be trained simply with observations. The approach can be applied for the correction of point-like as well as extended objects. Leveraging the linear image formation theory and a probabilistic approach to the blind deconvolution problem produces a physically-motivated loss function. The optimization of this loss function allows an end-to-end training of a machine learning model composed of three neural networks. As examples, we apply this procedure to the deconvolution of stellar data from the FastCam instrument and to solar extended data from the Swedish Solar Telescope. The analysis demonstrates that the proposed neural model can be successfully trained without supervision using observations only. It provides estimations of the instantaneous wavefronts, from which a corrected image can be found using standard deconvolution technniques. The network model is roughly three orders of magnitude faster than applying standard deconvolution based on optimization and shows potential to be used on real-time at the telescope.

arXiv comment11 pages, 4 figures, accepted for publication in A&A

References in corpus (8)

• Array Programming with NumPy
• CRISPRED: A data pipeline for the CRISP imaging spectropolarimeter
• Real-time multiframe blind deconvolution of solar images
• Accurate phase retrieval of complex point spread functions with deep residual neural networks
• Fast Point Spread Function Modeling with Deep Learning
• Blind deconvolution in astronomy with adaptive optics: the parametric marginal approach
• Image restoration of solar spectra
• Space variant deconvolution of galaxy survey images

Cited by in corpus (5)

• Instrument-To-Instrument translation: Instrumental advances drive restoration of solar observation series via deep learning
• Accelerating Multiframe Blind Deconvolution via Deep Learning
• torchmfbd: a flexible multi-object multi-frame blind deconvolution code
• Solar multi-object multi-frame blind deconvolution with a spatially variant convolution neural emulator
• Multi-frame blind deconvolution and phase diversity with statistical inclusion of uncorrected high-order modes