Classification of stars, galaxies, and quasars based on photometric multiband images

• Grant number: 22/11645-1
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: October 01, 2022
• End date: November 30, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Nina Sumiko Tomita Hirata
• Grantee: Gabriel Jacob Perin
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 15/22308-2 - Intermediate representations in Computational Science for knowledge discovery, AP.TEM
• Associated scholarship(s): 23/15047-4 - Model Merging for Large Language Model, BE.EP.IC

Abstract

The S-PLUS (Southern Photometric Local Universe Survey) project is collecting images of 12 bands of the southern hemisphere sky. In the context of this project, there is interest in the detection of quasars and other types of objects. In previous works, methods based on machine learning were developed to classify three types of objects (stars, galaxies, and quasars). One of the works used catalog data, including objects with missing data, while the other used images, without objects with missing data. The objective of this scientific initiation project is to compare these methods using a new data release, DR4, from S-PLUS. For this, the implementations of the methods will be reviewed and adapted to the new scenario and the evaluations will be carried out under experimental conditions directly comparable. The study will involve investigations into the treatment of missing data and pre-training of convolutional neural networks, to be used in image processing. Quantitative and qualitative comparisons of the two methods will be carried out to understand their differences.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

PERIN, GABRIEL J.; HIRATA, NINA S. T.. Few-shot Retinal Disease Classification on the Brazilian Multilabel Ophtalmological Dataset. 2024 37TH SIBGRAPI CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES, SIBGRAPI 2024, v. N/A, p. 6-pg., 2024-01-01. (23/15047-4, 22/11645-1, 22/15304-4)

HERPICH, F. R.; ALMEIDA-FERNANDES, F.; SCHWARZ, G. B. OLIVEIRA; LIMA, E. V. R.; NAKAZONO, L.; ALONSO-GARCIA, J.; FONSECA-FARIA, M. A.; SARTORI, M. J.; BOLUTAVICIUS, G. F.; DE SOUZA, G. FABIANO; et al. The Fourth S-PLUS Data Release: 12-filter photometry covering ∼3000 square degrees in the southern hemisphere. Astronomy & Astrophysics, v. 689, p. 25-pg., 2024-09-20. (23/10774-5, 24/03575-9, 18/20977-2, 21/10429-0, 21/11345-5, 22/00996-8, 21/09468-1, 22/11645-1, 23/14272-4, 18/21661-9, 20/00457-4, 19/01312-2, 21/12744-0, 21/08813-7)

PERIN, GABRIEL J.; CHEN, XUXI; LIU, SHUSEN; KAILKHURA, BHAVYA; WANG, ZHANGYANG; GALLAGHER, BRIAN. RankMean: Module-Level Importance Score for Merging Fine-tuned Large Language Models. FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS: ACL 2024, v. N/A, p. 7-pg., 2024-01-01. (23/15047-4, 22/11645-1, 22/15304-4)