Slide 1

Production, visualization and analysis of large volumes of remote sensing images modeled as multidimensional data cubes for the entire Brazilian territory.

New publication by the Brazil Data Cube team

Scholarship holder Lorena Santos, from the Brazil Data Cube project, published the article “Quality control and class noise reduction of satellite image time series” in the ISPRS Journal of Photogrammetry and Remote Sensing, together with her OTG advisors and project researchers, Karine Ferreira, Gilberto Camara and Michelle Picoli. “

The article is the result of her PHD at Brazil’s National Institute for Space Research on Applied Computing, and it shows the importance of using good quality samples when machine learning techniques are applied. The article proposes a method that uses self-organizing maps and Bayesian inference to reduce class noise through satellite images time series.

To see other publications by the BDC team, visit

Source: Brazil Data Cube Project.

  • Method for class noise reduction in satellite image time series reference data.
Brazil Data Cube - 2019 - 2024