12-02, 14:00–18:00 (America/Belem), Room Mangal das Garças (C Block)
This workshop will address the data products and software tools of the Brazil Data Cube Platform (http://brazildatacube.org/). The Brazil Data Cube (BDC) project is producing more than 2 petabytes of Analysis-Read Data (ARD) and multidimensional data cubes of satellite images Landsat-8/-9, Sentinel-2, CBERS-4/-4A and Amazonia for the entire Brazilian territory. Besides that, the BDC project is developing software tools to deal with big data sets, to extract image time series from Earth observation (EO) data cubes and to produce land use and land cover information using image time series and machine learning.
This workshop will address concepts of EO data cubes and satellite image time series analysis as well as promote hands-on activities using Python language for: (1) Discovering, accessing and viewing EO data cubes; (2) Extraction of satellite image time series from EO data cubes; (3) Analysis of satellite image time series; and (4) Extraction of land use and land cover trajectories. All software tools of the BDC platform will be demonstrated: BDCExplorer, TerraCollect, Satellite Image Time Series (SITS) R package and the web services, SpatioTemporal Asset Catalog (STAC), Web Time Series Service (WTSS) and Web Land Trajectory Service (WLTS).
The main goal of the BDC is to support environmental monitoring, land use and land cover applications, agricultural management, and other applications that require consistent and temporally structured EO satellite images and geospatial information. Image time series extracted from EO data cubes improve our understanding of environmental patterns and processes. Instead of selecting individual images from specific dates and comparing them, researchers can track change continuously. Satellite image time series analysis captures subtle changes in ecosystems and improves the quality of land classification.
BDC provides ARD and multidimensional data cubes of images from sensors onboard Landsat-8, Landsat-9, Sentinel-2, CBERS-4, CBERS-4A and AMAZONIA-1 satellites. Products like MOD13Q1 and MYD13Q1, which are derived from TERRA/MODIS and AQUA/MODIS satellite/sensor are also incorporated into BDC as data cubes. Using the same technologies to produce EO data cubes, BDC also produces Visualization Mosaics. BDC manages more than 2 petabytes of data, which brings big data challenges. Thus, the BDC platform also provides software tools to efficiently deal with these big EO data sets.
The software tools of the BDC platform that will be presented in this workshop are: (1) BDCExplorer: web portal to access, visualize and download EO data cubes and extract image times series; (2) TerraCollect: web application to collect and analyze land use and land cover samples, based on satellite image time series; (3) Satellite Image Time Series (SITS) R package for land use and land cover classification using satellite image times series and machine learning; and (4) Web services and R and Python clients. The web services include: (1) Spatio Temporal Asset Catalog (STAC); Web Time Series Service (WTSS) to extract image time series from EO data cubes; (3) Web Land Trajectory Service (WLTS) to extract land use and land cover trajectories from distinct data sets.
The speakers of this workshop are: Karine Reis Ferreira (INPE), Gilberto Ribeiro de Queiroz (INPE), Baggio Luiz de Castro e Silva (INPE) e Rennan de Freitas Bezerra Marujo (INPE).
This workshop will address the data products and software tools of the Brazil Data Cube Platform (http://brazildatacube.org/). It will address concepts of EO data cubes and satellite image time series analysis as well as promote hands-on activities using Python language for: (1) Discovering, accessing and viewing EO data cubes; (2) Extraction of satellite image time series from EO data cubes; (3) Analysis of satellite image time series; and (4) Extraction of land use and land cover trajectories. All software tools of the BDC platform will be demonstrated: BDCExplorer, TerraCollect, Satellite Image Time Series (SITS) R package and the web services, SpatioTemporal Asset Catalog (STAC), Web Time Series Service (WTSS) and Web Land Trajectory Service (WLTS).
For more information regarding Brazil Data Cube (BDC), see:
Ferreira, K.R.; Queiroz, G.R.; Vinhas, L.; Marujo, R.F.B.; Simoes, R.E.O.; Picoli, M.C.A.; Camara, G.; Cartaxo, R.; Gomes, V.C.F.; Santos, L.A.; Sanchez, A.H.; Arcanjo, J.S.; Fronza, J.G.; Noronha, C.A.; Costa, R.W.; Zaglia, M.C.; Zioti, F.; Korting, T.S.; Soares, A.R.; Chaves, M.E.D.; Fonseca, L.M.G. Earth Observation Data Cubes for Brazil: Requirements, Methodology and Products. Remote Sens. 2020, 12, 4033. https://doi.org/10.3390/rs12244033
For more information regarding the Web Time Series Service (WTSS), see:
VINHAS, L.; QUEIROZ, G. R.; FERREIRA, K. R.; C MARA, G. Web Services for Big Earth Observation Data. Revista Brasileira de Cartografia. 2017, 69, 5, 18. https://seer.ufu.br/index.php/revistabrasileiracartografia/article/view/44004
For more information regarding the Web Land Trajectory Service (WLTS), see:
Fabiana Zioti, Karine R. Ferreira, Gilberto R. Queiroz, Alana K. Neves, Felipe M. Carlos, Felipe C. Souza, Lorena A. Santos, Rolf E.O. Simoes, A platform for land use and land cover data integration and trajectory analysis, International Journal of Applied Earth Observation and Geoinformation. 2022, 106. https://doi.org/10.1016/j.jag.2021.102655
For more information regarding the R package Satellite Image Time Series (SITS), see:
Simoes, R.; Camara, G.; Queiroz, G.; Souza, F.; Andrade, P.R.; Santos, L.; Carvalho, A.; Ferreira, K. Satellite Image Time Series Analysis for Big Earth Observation Data. Remote Sens. 2021, 13, 2428. https://www.mdpi.com/2072-4292/13/13/2428
Karine Ferreira is PhD in Applied Computing and works at the National Institute for Space Research (INPE), Brazil, with research in Geoinformatics and coordinating technological innovation projects, such as Brazil Data Cube (http://brazildatacube.org/). She is professor of the Applied Computing Postgraduate Course at INPE and her main research topics is: representation, processing and analysis of spatiotemporal information, satellite image time series and big Earth observation data. She is a CNPq productivity scholarship in Technological Development and Innovative Extension.
PhD in Applied Computing and MSc in Remote Sensing by the Brazilian National Institute for Space Research (INPE) with a Bachelor in Computer Science by the Federal University of Lavras (UFLA). Works as Software Developer at the Brazil Data Cube project focused on Generation of Analysis Ready Data (ARD) and EO Data Cubes. Research: Digital Image Processing Algorithms, Image Segmentation, Artificial Intelligence, Data Mining, Atmosphere Correction and Radiometric corrections.
Baggio Castro é bacharel em Matemática com ênfase em Matemática Computacional pela Universidade Federal Fluminense. É mestre e doutorando em Computação Aplicada pelo Instituto Nacional de Pesquisas Espaciais. Trabalha no projeto Brazil Data Cube como cientista de dados geoespaciais e especialista em inteligência artificial aplicada a séries temporais de imagens de satélite para mapeamento de uso e cobertura da terra.
PhD student in Applied Computing at the Brazilian National Institute for Space Research (INPE). Master in Applied Computing and Technologist in Systems Analysis and Development. He has experience in Computer Science, with an emphasis on Geoinformatics. He is currently a researcher at the Foundation for Scientific and Technological Development in Health (FioTEC) and a full-stack developer on the Brazil Data Cube and HARMONIZE projects.