Abstract
Astronomers require efficient automated detection and classification pipelines when conducting large-scale surveys of the (optical) sky for variable and transient sources. Such pipelines are fundamentally important, as they permit rapid follow-up and analysis of those detections most likely to be of scientific value. We therefore present a deep learning pipeline based on the convolutional neural network architecture called MeerCRAB. It is designed to filter out the so called “bogus” detections from true astrophysical sources in the transient detection pipeline of the MeerLICHT telescope. Optical candidates are described using a variety of 2D images and numerical features extracted from those images. The relationship between the input images and the target classes is unclear, since the ground truth is poorly defined and often the subject of debate. This makes it difficult to determine which source of information should be used to train a classification algorithm. We therefore used two methods for labelling our data (i) thresholding and (ii) latent class model approaches. We deployed variants of MeerCRAB that employed different network architectures trained using different combinations of input images and training set choices, based on classification labels provided by volunteers. The deepest network worked best with an accuracy of 99.5% and Matthews correlation coefficient (MCC) value of 0.989. The best model was integrated to the MeerLICHT transient vetting pipeline, enabling the accurate and efficient classification of detected transients that allows researchers to select the most promising candidates for their research goals.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig7_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig8_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig9_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig10_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig11_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig12_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10686-021-09757-1/MediaObjects/10686_2021_9757_Fig13_HTML.png)
Similar content being viewed by others
Code Availability
MeerCRAB code and pre-trained models are available on Github at https://github.com/Zafiirah13/meercrab and on Zenodo at https://doi.org/10.5281/zenodo.4049943.
Notes
5 vetters labelled them as bogus and the other 5 as real.
References
Agarap, A.F.: Deep Learning using Rectified Linear Units (ReLU). arXiv:1803.08375 (2018)
Bellm, E.C., Kulkarni, S.R., Graham, M.J., et al.: The Zwicky Transient Facility: System Overview, Performance, and First Results. PASP 131, 018002 (2019)
Bertin, E.: Automated Morphometry with SExtractor and PSFEx. In: Evans, I.N., Accomazzi, A., Mink, D.J., Rots, A.H. (eds.) Astronomical Data Analysis Software and Systems XX, vol. 442, p. 435. Astronomical Society of the Pacific Conference Series (2011)
Bertin, E., Arnouts, S.: SExtractor: Software for source extraction. A&AS, vol 117 (1996)
Bloemen, S., Groot, P., Woudt, P., et al.: MeerLICHT and BlackGEM: custom-built telescopes to detect faint optical transients. In: SPIE, vol. 9906, p. 990664. Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series (2016)
Cabrera-Vives, G., Reyes, I., Förster, F., Estévez, P. A., Maureira, J.-C.: Deep-HiTS: Rotation Invariant Convolutional Neural Network for Transient Detection. ApJ 836, 97 (2017)
Chollet, F., et al.: Keras: The Python Deep Learning library (2018)
Drake, A.J., Djorgovski, S.G., Mahabal, A., et al.: First Results from the Catalina Real-Time Transient Survey. ApJ 696, 870 (2009)
Edwards, A.L.: Note on the “correction for continuity” in testing the significance of the difference between correlated proportions. Psychometrika, p. 13 (1948)
Formann, A.: K. Die latent-class-analyse, Einführung in Theorie und Anwendung (Beltz (1984)
Gaia Collaboration, Brown, A.G.A., Vallenari, A., et al.: Gaia Data Release 1. Summary of the astrometric, photometric, and survey properties. A&A 595, A2 (2016)
Gieseke, F., Bloemen, S., van den Bogaard, C., et al.: Convolutional neural networks for transient candidate vetting in large-scale surveys. MNRAS 472, 3101 (2017)
Groot, P.J.: The multi-colour dynamic Universe explored. Nature Astronomy 3, 1160 (2019)
Han, B., Yao, Q., Yu, X., et al.: Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labels. arXiv:1804.06872 (2018)
Hosenie, Z., Lyon, R., Stappers, B., Mootoovaloo, A., McBride, V.: Imbalance learning for variable star classification. MNRAS 493, 6050 (2020)
Hosenie, Z., Lyon, R.J., Stappers, B.W., Mootoovaloo, A.: Comparing Multiclass, Binary, and Hierarchical Machine Learning Classification schemes for variable stars. MNRAS 488, 4858 (2019)
Jonas, J., MeerKAT Team: The MeerKAT Radio Telescope. In: MeerKAT Science: On the Pathway to the SKA, p. 1 (2016)
Kaiser, N., Burgett, W., Chambers, K., et al.: The Pan-STARRS wide-field optical/NIR imaging survey. In: SPIE, vol .7733, p. 77330E. Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series (2010)
Keller, S.C., Schmidt, B.P., Bessell, M.S., et al.: The SkyMapper Telescope and The Southern Sky Survey. Publ. Astron. Soc. 826 Australia, vol. 24, p. 1 (2007)
Kingma, D.P., Ba, J.: Adam: A Method for Stochastic Optimization. arXiv:1412.6980 (2014)
Lang, D., Hogg, D.W., Mierle, K., Blanton, M., Roweis, S.: Astrometry.net: Blind Astrometric Calibration of Arbitrary Astronomical Images. AJ 139, 1782 (2010)
Lecun, Y., Haffner, P., Bottou, L., Bengio, Y.: Object Recognition with Gradient-Based Learning (1999)
Lin, H., Li, X., Zeng, Q.: Pulsar Candidate Sifting Using Multi-input Convolution Neural Networks. arXiv:2007.14843 (2020)
LSST Science Collaboration, Abell, P.A., Allison, J., et al.: LSST Science Book, Version 2.0. arXiv:0912.0201 (2009)
McCutcheon, A.L.: Latent class analysis, 64. Sage (1987)
McNemar, Q.: Note on the sampling error of the difference between correlated proportions or percentages. Psychometrika 12, 12 (1947)
Muthukrishna, D., Narayan, G., Mandel, K.S., Biswas, R., Hložek, R.: RAPID: Early Classification of Explosive Transients Using Deep Learning. PASP 131, 118002 (2019)
Paterson, K. In: Griffin, R.E. (ed.) Southern Horizons in Time-Domain Astronomy, vol. 339, pp. 203–203. IAU Symposium (2019)
Rau, A., Kulkarni, S.R., Law, N.M., et al.: Exploring the Optical Transient Sky with the Palomar Transient Factory. PASP 121, 1334 (2009)
Richards, J.W., Starr, D.L., Butler, N.R., et al.: On Machine-learned Classification of Variable Stars with Sparse and Noisy Time-series Data. ApJ 733, 10 (2011)
Shappee, B.J., Prieto, J.L., Grupe, D., et al.: The Man behind the Curtain: X-Rays Drive the UV through NIR Variability in the 2013 Active Galactic Nucleus Outburst in NGC 2617. ApJ, p. 788 (2014)
Vafaei Sadr, A., Vos, E.E., Bassett, B.A., et al.: DEEPSOURCE: point source detection using deep learning. MNRAS 484, 2793 (2019)
van Dokkum, P.G.: Cosmic-Ray Rejection by Laplacian Edge Detection. PASP 113, 1420 (2001)
Zackay, B., Ofek, E.O., Gal-Yam, A.: Proper Image Subtraction-Optimal Transient Detection, Photometry, and Hypothesis Testing. ApJ 830, 27 (2016)
Zhang, C., Bengio, S., Hardt, M., Recht, B., Vinyals, O.: Understanding deep learning requires rethinking generalization. arXiv:1611.03530 (2016)
Acknowledgements
We thank the referee for useful comments and suggestions for improving this paper. We would like to thank the people who gave up their time to do the vetting of the sample: Laura Driessen, Naomi titus, Mark Beijer, Nadia Blagorodnova, Joris Kersten, David Modiano and Roque Ruiz Carmona, without whose effort this work would not have been possible. We would like to also thank Arrykrishna Mootoovaloo and Fabian Gieseke for useful discussion. The MeerLICHT consortium is a partnership between Radboud University, the University of Cape Town, the Netherlands Organisation for Scientific Research (NWO), the South African Astronomical Observatory (SAAO), the University of Oxford, the University of Manchester and the University of Amsterdam, in association with and, partly supported by, the South African Radio Astronomy Observatory (SARAO), the European Research Council and the Netherlands Research School for Astronomy (NOVA).
Funding
ZH acknowledges support from the UK Newton Fund as part of the Development in Africa with Radio Astronomy (DARA) Big Data project delivered via the Science & Technology Facilities Council (STFC). BWS acknowledges funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. 694745). PJG and SDW are supported by NRF SARChI Grant 111692.
Author information
Authors and Affiliations
Corresponding author
Additional information
Availability of Data and Material
Data will be available upon request.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Hosenie, Z., Bloemen, S., Groot, P. et al. MeerCRAB: MeerLICHT classification of real and bogus transients using deep learning. Exp Astron 51, 319–344 (2021). https://doi.org/10.1007/s10686-021-09757-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10686-021-09757-1