Eliminating artefacts in polarimetric images using deep learning

Paranjpye, D. and Mahabal, A. and Ramaprakash, A. N. and Panopoulou, G. V. and Cleary, K. and Readhead, A. C. S. and Blinov, D. and Tassis, K. (2020) Eliminating artefacts in polarimetric images using deep learning. Monthly Notices of the Royal Astronomical Society, 491 (4). pp. 5151-5157. ISSN 0035-8711. doi:10.1093/mnras/stz3250. https://resolver.caltech.edu/CaltechAUTHORS:20200227-125520786

	PDF - Published Version See Usage Policy. 3MB
	PDF - Submitted Version See Usage Policy. 1MB

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20200227-125520786

Abstract

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.

Item Type:

Article

Related URLs:

URL	URL Type	Description
https://doi.org/10.1093/mnras/stz3250	DOI	Article
https://arxiv.org/abs/1911.08327	arXiv	Discussion Paper

ORCID:

Author	ORCID
Mahabal, A.	0000-0003-2242-0244
Panopoulou, G. V.	0000-0001-7482-5759
Readhead, A. C. S.	0000-0001-9152-961X
Tassis, K.	0000-0002-8831-2038

Additional Information:

© 2019 The Author(s) Published by Oxford University Press on behalf of the Royal Astronomical Society. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model). Accepted 2019 November 19. Received 2019 November 9; in original form 2019 October 4. Published: 28 November 2019. The work has been funded by the National Science Foundation under the NSF grant (161547). AM acknowledges support from the NSF (1640818, AST-1815034) and IUSSTF (JC-001/2017). KT acknowledges support from the European Research Council under the European Union’s Horizon 2020 research and innovation program, under grant agreement no. 771282.

Group:

Astronomy Department

Funders:

Funding Agency	Grant Number
NSF	AST-161547
NSF	OAC-1640818
NSF	AST-1815034
Indo-US Science and Technology Forum	JC-001/2017
European Research Council (ERC)	771282

Subject Keywords:

deep learning – image classication – artefect detection – polarmetry

Issue or Number:

DOI:

10.1093/mnras/stz3250

Record Number:

CaltechAUTHORS:20200227-125520786

Persistent URL:

https://resolver.caltech.edu/CaltechAUTHORS:20200227-125520786

Official Citation:

D Paranjpye, A Mahabal, A N Ramaprakash, G V Panopoulou, K Cleary, A C S Readhead, D Blinov, K Tassis, Eliminating artefacts in polarimetric images using deep learning, Monthly Notices of the Royal Astronomical Society, Volume 491, Issue 4, February 2020, Pages 5151–5157, https://doi.org/10.1093/mnras/stz3250

Usage Policy:

No commercial reproduction, distribution, display or performance rights in this work are provided.

ID Code:

101619

Collection:

CaltechAUTHORS

Deposited By:

Tony Diaz

Deposited On:

27 Feb 2020 21:08

Last Modified:

16 Nov 2021 18:04

Repository Staff Only: item control page