Brazil Data Cube is a project that is being developed by Brazil’s National Institute for Space Research (INPE ), since January 2019, that aims to create multidimensional data cubes of analysis-ready from medium-resolution Earth observation images for all Brazilian territory and to generate land use and cover information from these data cubes using machine learning and satellite image time series analysis.
The Brazil Data Cube project is part of the “Environmental Monitoring of Brazilian Biomes project“, funded by the Amazon Fund through the financial collaboration of the Brazilian Development Bank (BNDES) and the Foundation for Science, Technology and Space Applications (FUNCATE) no. 17.2.0536.1. Brazil Data Cube is the successor of the research project e-sensing, funded by FAPESP (Fapesp 2014/08398-6).
The Brazil Data Cube project has four main objectives:
1. Create analysis-ready data sets from medium-resolution remote sensing images (10 to 30 meters) for all Brazilian territory, including images from the Earth observation satellites Landsat, CBERS and Sentinel;
2. Model these analysis-ready data sets as multidimensional data cubes with three or more dimensions that include space, time and properties;
3. Use, propose and develop novel methods and big data technologies to store and process these data cube sets and to analyze and extract land use and cover information from them using satellite image time series analysis, machine learning and image processing procedures;
4. Create land use and cover information for Brazil using the data cubes and methods developed in this project.