http://arxiv.org/abs/2012.05947
With the advent of interferometric instruments with 4 telescopes at the VLTI and 6 telescopes at CHARA, the scientificpossibility arose to routinely obtain milli-arcsecond scale images of the observed targets. Such an image reconstructionprocess is typically performed in a Bayesian framework where the function to minimize is made of two terms: the datalikelihood and the Bayesian prior. This prior should be based on our prior knowledge of the observed source. Up to now,this prior was chosen from a set of generic and arbitrary functions, such as total variation for example. Here, we present animage reconstruction framework using generative adversarial networks where the Bayesian prior is defined using state-of-the-art radiative transfer models of the targeted objects. We validate this new image reconstruction algorithm on syntheticdata with added noise. The generated images display a drastic reduction of artefacts and allow a more straightforwardastrophysical interpretation. The results can be seen as a first illustration of how neural networks can provide significantimprovements to the image reconstruction post processing of a variety of astrophysical sources.
R. Claes, J. Kluska, H. Winckel, et. al.
Mon, 14 Dec 20
49/74
Comments: SPIE Astronomical Telescopes + Instrumentation 2020 conference Paper No. 11446-110
You must be logged in to post a comment.