Deep Learning assessment of galaxy morphology in S-PLUS Data Release 1

Autor(es):Mostrar menos -

Bom, C. R. ^{[1, 2]} ; Cortesi, A. ^[3] ; Lucatelli, G. ^[4] ; Dias, L. O. ^[1] ; Schubert, P. ^[1] ; Oliveira Schwarz, G. B. ^[5] ; Cardoso, N. M. ^[6] ; Lima, E. V. R. ^[4] ; de Oliveira, C. Mendes ^[4] ; Sodre, L. ^[4] ; Smith Castelli, V, A. ; Ferrari, F. ^[7] ; Damke, G. ^[8] ; Overzier, R. ^{[9, 4]} ; Kanaan, A. ^[10] ; Ribeiro, T. ^[11] ; Schoenell, W. ^[12]

Afiliação do(s) autor(es):Mostrar menos -

^[1] Ctr Brasileiro Pesquisas Fis, Rua Dr Xavier Sigaud 150, BR-22290180 Rio De Janeiro, RJ - Brazil

^[2] Ctr Fed Educ Tecnol Celso Suckow Fonseca, Rodovia Mario Covas, Lote 12, BR-23810000 Itaguai, RJ - Brazil

^[3] Univ Fed Rio de Janeiro, Valongo Observ, Ladeira Pedro Antonio 43, BR-20080090 Rio De Janeiro, RJ - Brazil

^[4] Univ Sao Paulo, IAG, Rua Mato 1225, BR-05508090 Sao Paulo, SP - Brazil

^[5] Univ Presbiteriana Mackenzie, R Consolacao 930, BR-01302907 Sao Paulo, SP - Brazil

^[6] Univ Sao Paulo, Escola Politecn, Av Prof Luciano Gualberto, Travessa Politecn 380, BR-05508010 Sao Paulo, SP - Brazil

^[7] Univ Fed Rio Grande IMEF FURG, Inst Matemat Estat & Fis, Av Italia Km 8, BR-96201900 Rio Grande, RS - Brazil

^[8] Univ La Serena, Inst Invest Multidisciplinar Ciencia & Tecnol, Raul Bitran 1305, La Serena - Chile

^[9] Observ Nacl, Rua Gen Jose Cristino 77, BR-20921400 Rio De Janeiro, RJ - Brazil

^[10] Univ Fed Santa Catarina, Dept Fis, BR-88040900 Florianopolis, SC - Brazil

^[11] Univ Fed Rio Grande do Sul UFRGS, Inst Fis, Dept Astron, POB 15051, Av Bento Goncalves 9500, Porto Alegre, RS - Brazil

^[12] Natl Opt Astron Observ, POB 26732, Tucson, AZ 85726 - USA

• Tipo de documento: Artigo Científico
• Fonte: Monthly Notices of the Royal Astronomical Society; v. 507, n. 2, p. 1937-1955, OCT 2021.
• Citações Web of Science: 0

Resumo

The morphological diversity of galaxies is a relevant probe of galaxy evolution and cosmological structure formation, but the classification of galaxies in large sky surveys is becoming a significant challenge. We use data from the Stripe-82 area observed by the Southern Photometric Local Universe Survey (S-PLUS) in 12 optical bands, and present a catalogue of the morphologies of galaxies brighter than r = 17 mag determined both using a novel multiband morphometric fitting technique and Convolutional Neural Networks (CNNs) for computer vision. Using the CNNs, we find that, compared to our baseline results with three bands, the performance increases when using 5 broad and 3 narrow bands, but is poorer when using the full 12 band S-PLUS image set. However, the best result is still achieved with just three optical bands when using pre-trained network weights from an ImageNet data set. These results demonstrate the importance of using prior knowledge about neural network weights based on training in unrelated, extensive data sets, when available. Our catalogue contains 3274 galaxies in Stripe-82 that are not present in Galaxy Zoo 1 (GZ1), and we also provide our classifications for 4686 galaxies that were considered ambiguous in GZ1. Finally, we present a prospect of a novel way to take advantage of 12 band information for morphological classification using morphometric features, and we release a model that has been pre-trained on several bands that could be adapted for classifications using data from other surveys. The morphological catalogues are publicly available. (AU)

• Processo FAPESP: 19/26492-3 - Ciência com o telescópio robótico brasileiro
• Beneficiário: Claudia Lucia Mendes de Oliveira
• Modalidade de apoio: Auxílio à Pesquisa - Projetos Especiais
• Processo FAPESP: 09/54202-8 - EMU: aquisição de um telescópio robótico para a comunidade astronômica brasileira
• Beneficiário: Claudia Lucia Mendes de Oliveira
• Modalidade de apoio: Auxílio à Pesquisa - Programa Equipamentos Multiusuários
• Processo FAPESP: 19/10923-5 - Ferramentas de inteligência artificial para grandes levantamentos de galáxias
• Beneficiário: Laerte Sodré Junior
• Modalidade de apoio: Auxílio à Pesquisa - Regular
• Processo FAPESP: 19/11910-4 - Explorando dados do SAM-FP e T80-Sul
• Beneficiário: Claudia Lucia Mendes de Oliveira
• Modalidade de apoio: Auxílio à Pesquisa - Regular