/IDeepS

Project IDeepS: Image and data classification via Deep neural networks for multiple domainS

Primary LanguagePythonGNU General Public License v3.0GPL-3.0

Project IDeepS

The project Classificação de imagens e dados via redes neurais profundas para múltiplos domínios (Image and data classification via Deep neural networks for multiple domainS - IDeepS) is a continuation of a previous project and whose current objective is to propose recommendations/suggestions for the best deep neural network (DNN) models to be used for the remote sensing (which includes aerial images obtained by unmanned aerial vehicles (UAVs), airplanes, as well as images obtained by satellites), astrophysics, and health domains. Image classification will be the main computer vision task considered, but other tasks will be evaluated, also taking into account the greater diversity of distinct data.

The IDeepS project is supported by the Laboratório Nacional de Computação Científica (LNCC/MCTI, Brazil) via resources of the SDumont supercomputer. Researchers, professors and post-graduate students from the following organisations are involved in the project: Instituto Nacional de Pesquisas Espaciais (INPE), Instituto de Estudos Avançados (IEAv), Universidade Federal de São Paulo - Campus São José dos Campos (UNIFESP), Universidade Federal de São Carlos - Campus Sorocaba (UFSCar), and Universidade Estadual Paulista Júlio de Mesquita Filho - Campus Bauru (UNESP).

Publications

✅ M. S. Miranda, V. A. Santiago Júnior, T. S. Körting, E. C. S. Monteiro, and J. Q. Silva. AI4LUC: Deep learning and automated mask labelling to support land use and land cover mapping in the Cerrado biome. Remote Sensing Letters, v. 15, n. 8, p. 850 - 860, 2024. Access here.

Team

Member Organisation
Adriano Pereira Almeida INPE
Alan James Peixoto Calheiros INPE
Álvaro Luiz Fazenda UNIFESP
André Estevam Costa Oliveira INPE
Bruno Nardi de Carvalho Dantas ITA
Caio Eduardo Dias INPE
Daniel Augusto de Sousa Mendes INPE
Eduardo Bouhid Neto UNIFESP
Elcio Hideiti Shiguemori IEAv
Gabriel Lino Garcia UNESP
Hércules Carlos Dos Santos Pereira INPE
Hugo Resende UNIFESP
João Paulo Papa UNESP
Jurandy Gomes de Almeida Junior UFSCar
Mateus de Souza Miranda INPE
Matheus Corrêa Domingos INPE
Nathan Augusto Zacarias Xavier ITA
Rafael Marinho de Andrade INPE
Reinaldo Roberto Rosa INPE
Samuel Felipe dos Santos UNIFESP
Thales Sehn Körting INPE
Valdivino Alexandre de Santiago Júnior (Coordinator) INPE

SDumont User Manual: Deep Learning

Directives and suggestions on how one can perform the setup and run deep learning (DL) applications in the SDumont supercomputer are presented here.

Author

Valdivino Alexandre de Santiago Júnior

Licence

This project is licensed under the GNU GENERAL PUBLIC LICENSE, Version 3 (GPLv3) - see the LICENSE.md file for details.

Cite

Please cite this repository if you use it as:

V. A. Santiago Júnior. Project IDeepS, 2024. Acessed on: date of access. Available at: https://github.com/vsantjr/IDeepS.