Euclid preparation XXXIII. Characterization of convolutional neural networks for the identification of galaxy-galaxy strong-lensing events
Year: 2025
Type: article
Abstract: Forthcoming imaging surveys will increase the number of known galaxy-scale strong lenses by several orders of magnitude. For this to happen, images of billions of galaxies will have to be inspected to identify potential candidates. In this context, deep-learning techniques are particularly suitable for finding patterns in large data sets, and convolutional neural networks (CNNs) in particular can efficiently process large volumes of images. We assess and compare the performance of three network architectures in the classification of strong-lensing systems on the basis of their morphological characteristics. In particular, we implemented a classical CNN architecture, an inception network, and a residual network. We trained and tested our networks on different subsamples of a data set of 40 000 mock images whose characteristics were similar to those expected in the wide survey planned with the ESA mission Euclid, gradually including larger fractions of faint lenses. We also evaluated the importance of adding information about the color difference between the lens and source galaxies by repeating the same training on single- and multiband images. Our models find samples of clear lenses with greater than or similar to 90% precision and completeness. Nevertheless, when lenses with fainter arcs are included in the training set, the performance of the three models deteriorates with accuracy values of similar to 0.87 to similar to 0.75, depending on the model. Specifically, the classical CNN and the inception network perform similarly in most of our tests, while the residual network generally produces worse results. Our analysis focuses on the application of CNNs to high-resolution space-like images, such as those that the Euclid telescope will deliver. Moreover, we investigated the optimal training strategy for this specific survey to fully exploit the scientific potential of the upcoming observations. We suggest that training the networks separately on lenses with different morphology might be needed to identify the faint arcs. We also tested the relevance of the color information for the detection of these systems, and we find that it does not yield a significant improvement. The accuracy ranges from similar to 0.89 to similar to 0.78 for the different models. The reason might be that the resolution of the Euclid telescope in the infrared bands is lower than that of the images in the visual band.
Source: UTUPub (University of Turku)
Authors: Leuzzi L., Meneghetti M., Angora G., Metcalf R. B., Moscardini L., Rosati P., Bergamini P., Calura F., Clement B., Gavazzi R., Gentile F., Lochner M., Grillo C., Vernardos G., Aghanim N., Amara A., Amendola L., Auricchio N., Bodendorf C., Bonino D., Branchini E., Brescia M., Brinchmann J., Camera S., Capobianco V., Carbone C., Carretero J., Castellano M., Cavuoti S., Cimatti A., Cledassou R., Congedo G., Conselice C. J., Conversi L., Copin Y., Corcione L., Courbin F., Cropper M., Da Silva A., Degaudenzi H., Dinis J., Dubath F., Dupac X., Dusini S., Farrens S., Ferriol S., Frailis M., Franceschi E., Fumana M., Galeotta S., Gillis B., Giocoli C., Grazian A., Grupp F., Guzzo L., Haugan S. V. H., Holmes W., Hormuth F., Hornstrup A., Hudelot P., Jahnke K., Kuemmel M., Kermiche S., Kiessling A., Kitching T., Kunz M., Kurki-Suonio H., Lilje P. B., Lloro I., Maiorano E., Mansutti O., Marggraf O., Markovic K., Marulli F., Massey R., Medinaceli E., Mei S., Melchior M., Mellier Y., Merlin E., Meylan G., Moresco M., Munari E., Niemi S. -M., Nightingale J. W., Nutma T., Padilla C., Paltani S., Pasian F., Pedersen K., Pettorino V., Pires S., Polenta G., Poncet M., Raison F., Renzi A., Rhodes J., Riccio G., Romelli E., Roncarelli M., Rossetti E., Saglia R., Sapone D., Sartoris B., Schneider P., Secroun A., Seidel G., Serrano S., Sirignano C., Sirri G., Stanco L., Tallada-Crespi P., Taylor A. N., Tereno I., Toledo-Moreo R., Torradeflot F., Tutusaus I., Valenziano L., Vassallo T., Wang Y., Weller J., Zamorani G., Zoubian J., Andreon S., Bardelli S., Boucaud A., Bozzo E., Colodro-Conde C., Di Ferdinando D., Farina M., Farinelli R., Gracia-Carpio J., Keihaenen E., Lindholm V., Maino D., Mauri N., Neissner C., Schirmer M., Scottez V., Tenti M., Tramacere A., Veropalumbo A., Zucca E., Akrami Y., Allevato V., Baccigalupi C., Ballardini M., Bernardeau F., Biviano A., Borgani S., Borlaff A. S., Bretonniere H., Burigana C., Cabanac R., Cappi A., Carvalho C. S., Casas S., Castignani G., Castro T., Chambers K. C., Cooray A. R., Coupon J., Courtois H. M., Davini S., de la Torre S., De Lucia G., Desprez G., Di Domizio S., Dole H., Escartin Vigo J. A., Escoffier S., Ferrero I., Gabarra L., Ganga K., Garcia-Bellido J., Gaztanaga E., George K., Gozaliasl G., Hildebrandt H., Hook I., Huertas-Company M., Joachimi B., Kajava J. J. E., Kansal V., Kirkpatrick C. C., Legrand L., Loureiro A., Magliocchetti M., Mainetti G., Maoli R., Martinelli M., Martinet N., Martins C. J. A. P., Matthew S., Maurin L., Monaco P., Morgante G., Nadathur S., Nucita A. A., Patrizii L., Popa V., Porciani C., Potter D., Pontinen M., Reimberg P., Sanchez A. G., Sakr Z., Schneider A., Sereno M., Simon P., Spurio Mancini A., Stadel J., Steinwagner J., Teyssier R., Valiviita J., Viel M., Zinchenko I. A., Dominguez Sanchez H.
Institutions: Institut de Physique des 2 Infinis de Lyon, École Polytechnique Fédérale de Lausanne, Centre National de la Recherche Scientifique, Aix-Marseille Université, Institut d'Astrophysique de Paris, Sorbonne Université, University of Bologna, South African Radio Astronomy Observatory, University of the Western Cape, Université Paris-Saclay, Institut d'Astrophysique Spatiale, University of Portsmouth, Heidelberg University, Max Planck Institute for Extraterrestrial Physics, Osservatorio Astrofisico di Torino, Istituto Nazionale di Fisica Nucleare, Sezione di Genova, University of Genoa, University of Naples Federico II, Universidade do Porto, Institute of Astrophysics and Space Sciences, Centro de Astrofísica da Universidade do Porto, National Institute for Astrophysics, Institute for High Energy Physics, Port d'Informació Científica, Astronomical Observatory of Rome, Istituto Nazionale di Fisica Nucleare, Sezione di Napoli, Centre National d'Études Spatiales, Institut National de Physique Nucléaire et de Physique des Particules, Royal Observatory, University of Edinburgh, University of Manchester, European Space Agency, European Space Astronomy Centre, European Space Research Institute, Lyon 1 Université, Université de Lyon, University College London, University of Lisbon, University of Geneva, Istituto Nazionale di Fisica Nucleare, Sezione di Padova, Université Paris Cité, Commissariat à l'Énergie Atomique et aux Énergies Alternatives, Astrophysique, Instrumentation et Modélisation, Trieste Astronomical Observatory, Osservatorio Astronomico di Padova, Ludwig-Maximilians-Universität München, Brera Astronomical Observatory, Istituto Nazionale di Fisica Nucleare, Sezione di Milano, University of Oslo, California Institute of Technology, Jet Propulsion Laboratory, von Hoerner & Sulger (Germany), Technical University of Denmark, Max Planck Institute for Astronomy, Centre de physique des particules de Marseille
Language: English
Subfield: Astronomy and Astrophysics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: green
Funders: National Aeronautics and Space Administration, European Space Agency, Ministero dell’Istruzione, dell’Università e della Ricerca, Agenzia Spaziale Italiana, Ministerio de Ciencia e Innovación, National Astronomical Observatory of Japan, Agenția Spațială Română, Norsk Romsenter, Institut de Física d'Altes Energies, Fundação para a Ciência e a Tecnologia, Staatssekretariat für Bildung, Forschung und Innovation, Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas, Integrated Electronics Engineering Center, Binghamton University
OpenAlex ID: W7137621647
FWCI:
Cites:
Cited by:
Related to: