Cosmic Ray rejection with attention augmented deep learning

Bhavanam, S.R. and Channappayya, S.S. and Srijith, P.K. and Desai, S. (2022) Cosmic Ray rejection with attention augmented deep learning. Astronomy and Computing, 40. pp. 1-17. ISSN 2213-1337

[img] Text
Astronomy_and_Computing.pdf - Published Version
Available under License Creative Commons Attribution.

Download (3MB)


Cosmic Ray (CR) hits are the major contaminants in astronomical imaging and spectroscopic observations involving solid-state detectors. Correctly identifying and masking them is a crucial part of the image processing pipeline, since it may otherwise lead to spurious detections. For this purpose, we have developed and tested a novel Deep Learning based framework for the automatic detection of CR hits from astronomical imaging data from two different imagers: Dark Energy Camera (DECam) and Las Cumbres Observatory Global Telescope (LCOGT). We considered two baseline models namely deepCR and Cosmic-CoNN, which are the current state-of-the-art learning based algorithms that were trained using Hubble Space Telescope (HST) ACS/WFC and LCOGT Network images respectively. We have experimented with the idea of augmenting the baseline models using Attention Gates (AGs) to improve the CR detection performance. We have trained our models on DECam data and demonstrate a consistent marginal improvement by adding AGs in True Positive Rate (TPR) at 0.01% False Positive Rate (FPR) and Precision at 95% TPR over the aforementioned baseline models for the DECam dataset. We demonstrate that the proposed AG augmented models provide significant gain in TPR at 0.01% FPR when tested on previously unseen LCO test data having images from three distinct telescope classes. Furthermore, we demonstrate that the proposed baseline models with and without attention augmentation outperform state-of-the-art models such as Astro-SCRAPPY, Maximask (that is trained natively on DECam data) and pre-trained ground-based Cosmic-CoNN. This study demonstrates that the AG module augmentation enables us to get a better deepCR and Cosmic-CoNN models and to improve their generalization capability on unseen data. © 2022 Elsevier B.V.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Channappayya, Sumohana S.
Srijith, P K
Desai, Shantanu
Item Type: Article
Additional Information: SR was supported by Tata Consultancy Services (TCS) and Department of Science and Technology - Interdisciplinary Cyber Physical Systems (DST-ICPS) (under the grant T-641). We are grateful to the anonymous referees for several constructive comments and feedback on our manuscript. This project used data obtained with the Dark Energy Camera (DECam), which was constructed by the Dark Energy Survey (DES) collaborating institutions: Argonne National Lab, University of California Santa Cruz, University of Cambridge, Centro de Investigaciones Energeticas, Medioambientales y Tecnologicas-Madrid, University of Chicago, University College London, DES-Brazil consortium, University of Edinburgh, ETH-Zurich, University of Illinois at Urbana-Champaign, Institut de Ciencies de l'Espai, Institut de Fisica d'Altes Energies, Lawrence Berkeley National Lab, Ludwig-Maximilians Universitat, University of Michigan, National Optical Astronomy Observatory, University of Nottingham, Ohio State University, University of Pennsylvania, University of Portsmouth, SLAC National Lab, Stanford University, University of Sussex, and Texas A&M University. Funding for DES, including DECam, has been provided by the U.S. Department of Energy, National Science Foundation, Ministry of Education and Science (Spain), Science and Technology Facilities Council (UK), Higher Education Funding Council (England), National Center for Supercomputing Applications, Kavli Institute for Cosmological Physics, Financiadora de Estudos e Projetos, Fundacao Carlos Chagas Filho de Amparo a Pesquisa, Conselho Nacional de Desenvolvimento Científico e Tecnologico and the Ministerio da Ciencia e Tecnologia (Brazil), the German Research Foundation-sponsored cluster of excellence “Origin and Structure of the Universe” and the DES collaborating institutions. This work also makes use of observations from the Las Cumbres Observatory Global Telescope Network (Brown et al. 2013) having 25 telescopes at seven sites around the world. Astropy (Robitaille et al. 2013), LACosmic (Van Dokkum, 2001), Astro-SCRAPPY (McCully and Tewes, 2019), Numpy (Harris et al. 2020), Scipy (Virtanen et al. 2020), Matplotlib (Hunter, 2007), Jupyter (Kluyver et al. 2016), scikit-image (Van der Walt et al. 2014), SExtractor (Bertin and Arnouts, 1996), and Pytorch (Paszke et al. 2019).
Uncontrolled Keywords: Attention gates; CCD observation; Cosmic rays; Image processing; Neural networks
Subjects: Computer science
Electrical Engineering
Divisions: Department of Computer Science & Engineering
Department of Electrical Engineering
Department of Physics
Depositing User: . LibTrainee 2021
Date Deposited: 13 Aug 2022 12:51
Last Modified: 13 Aug 2022 12:51
Publisher URL:
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 10169 Statistics for this ePrint Item