AutoSourceID-Classifier

Star-galaxy classification using a convolutional neural network with spatial information

¹ Department of Astrophysics/IMAPP, Radboud University, PO Box 9010, 6500 GL Nijmegen, The Netherlands
e-mail: f.stoppa@astro.ru.nl
² Center for Astrophysics and Cosmology, University of Nova Gorica, Vipavska 13, 5000 Nova Gorica, Slovenia
³ High Energy Physics/IMAPP, Radboud University, PO Box 9010, 6500 GL Nijmegen, The Netherlands
⁴ Nikhef, Science Park 105, 1098 XG Amsterdam, The Netherlands
⁵ Instituto de Física Corpuscular, IFIC-UV/CSIC, Valencia, Spain
⁶ Department of Mathematics/IMAPP, Radboud University, PO Box 9010, 6500 GL Nijmegen, The Netherlands
⁷ SRON, Netherlands Institute for Space Research, Sorbonnelaan 2, 3584 CA Utrecht, The Netherlands
⁸ Institute of Astronomy, KU Leuven, Celestijnenlaan 200D, 3001 Leuven, Belgium
⁹ Institute for Fundamental Physics of the Universe, Via Beirut 2, 34151 Trieste, Italy
¹⁰ Department of Astronomy and Inter-University Institute for Data Intensive Astronomy, University of Cape Town, Private Bag X3, Rondebosch 7701, South Africa
¹¹ South African Astronomical Observatory, PO Box 9, Observatory, Cape Town 7935, South Africa
¹² Dipartimento di Fisica, Universitá di Trieste, 34127 Trieste, Italy
¹³ Istituto Nazionale di Fisica Nucleare, Sezione di Trieste, 34127 Trieste, Italy
¹⁴ Erlangen Centre for Astroparticle Physics, Nikolaus-Fiebiger-Str. 2, Erlangen 91058, Germany

Received: 26 July 2023
Accepted: 17 October 2023

Abstract

Aims. Traditional star-galaxy classification techniques often rely on feature estimation from catalogs, a process susceptible to introducing inaccuracies, thereby potentially jeopardizing the classification’s reliability. Certain galaxies, especially those not manifesting as extended sources, can be misclassified when their shape parameters and flux solely drive the inference. We aim to create a robust and accurate classification network for identifying stars and galaxies directly from astronomical images.

Methods. The AutoSourceID-Classifier (ASID-C) algorithm developed for this work uses 32x32 pixel single filter band source cutouts generated by the previously developed AutoSourceID-Light (ASID-L) code. By leveraging convolutional neural networks (CNN) and additional information about the source position within the full-field image, ASID-C aims to accurately classify all stars and galaxies within a survey. Subsequently, we employed a modified Platt scaling calibration for the output of the CNN, ensuring that the derived probabilities were effectively calibrated, delivering precise and reliable results.

Results. We show that ASID-C, trained on MeerLICHT telescope images and using the Dark Energy Camera Legacy Survey (DECaLS) morphological classification, is a robust classifier and outperforms similar codes such as SourceExtractor. To facilitate a rigorous comparison, we also trained an eXtreme Gradient Boosting (XGBoost) model on tabular features extracted by SourceExtractor. While this XGBoost model approaches ASID-C in performance metrics, it does not offer the computational efficiency and reduced error propagation inherent in ASID-C’s direct image-based classification approach. ASID-C excels in low signal-to-noise ratio and crowded scenarios, potentially aiding in transient host identification and advancing deep-sky astronomy.