Eliminating artefacts in polarimetric images using deep learning

al., et (author)

Panopoulou, Georgia, 1989 (author): California Institute of Technology (Caltech)

(creator_code:org_t)

2019-11-28
2020
English.
In: Monthly Notices of the Royal Astronomical Society. - : Oxford University Press (OUP). - 0035-8711 .- 1365-2966.

Related links:: https://authors.libr...; show more...; https://doi.org/10.1...; https://research.cha...; show less...

Abstract Subject headings

Polarization measurements done using Imaging Polarimeters such as the Robotic Polarimeter are very sensitive to the presence of artefacts in images. Artefacts can range from internal reflections in a telescope to satellite trails that could contaminate an area of interest in the image. With the advent of wide-field polarimetry surveys, it is imperative to develop methods that automatically flag artefacts in images. In this paper, we implement a Convolutional Neural Network to identify the most dominant artefacts in the images. We find that our model can successfully classify sources with 98 per cent true positive and 97 per cent true negative rates. Such models, combined with transfer learning, will give us a running start in artefact elimination for near-future surveys like WALOP.

NATURVETENSKAP -- Fysik -- Astronomi, astrofysik och kosmologi (hsv//swe)
NATURAL SCIENCES -- Physical Sciences -- Astronomy, Astrophysics and Cosmology (hsv//eng)
NATURVETENSKAP -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)
NATURAL SCIENCES -- Computer and Information Sciences -- Computer Sciences (hsv//eng)

About the subject

NATURAL SCIENCES: NATURAL SCIENCES; and Physical Science ...; and Astronomy Astrop ...

NATURAL SCIENCES: NATURAL SCIENCES; and Computer and Inf ...; and Computer Science ...

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.