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

Stoppa F, Bhattacharyya S, Ruiz De Austri R, Vreeswijk P, Caron S, Zaharijas G, Bloemen S, Principe G, Malyshev D, Vodeb V, Groot PJ, Cator E, Nelemans G (2023)

Publication Type: Journal article

Publication year: 2023

Journal

Astronomy & Astrophysics EDP Sciences

Book Volume: 680

Article Number: A109

DOI: 10.1051/0004-6361/202347576

Abstract

Aims. Traditional star-galaxy classification techniques often rely on feature estimation from catalogs, a process susceptible to introducing inaccuracies, thereby potentially jeopardizing the classificationa'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-Ca'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.

Authors with CRIS profile

Dmitry Malyshev Lehrstuhl für Physik

Involved external institutions

Radboud University Nijmegen NL Netherlands (NL) University of Nova Gorica / Univerza v Novi Gorici SI Slovenia (SI) Institute for Corpuscular Physics / Instituto de Física Corpuscular (IFIC) ES Spain (ES) Università degli Studi di Trieste IT Italy (IT)

How to cite

APA:

Stoppa, F., Bhattacharyya, S., Ruiz De Austri, R., Vreeswijk, P., Caron, S., Zaharijas, G.,... Nelemans, G. (2023). AutoSourceID-Classifier: Star-galaxy classification using a convolutional neural network with spatial information. Astronomy & Astrophysics, 680. https://doi.org/10.1051/0004-6361/202347576

MLA:

Stoppa, F., et al. "AutoSourceID-Classifier: Star-galaxy classification using a convolutional neural network with spatial information." Astronomy & Astrophysics 680 (2023).

BibTeX: Download