2024-10-03, 11:30–12:15, Wodak Room (IIASA)
This hands-on workshop will present the use of big data analytics to work with data available at the Open GEO Hub cloud service
The workshop will focus on the analysis of image time series extracted from big Earth observation data cubes available at the Open GEO Hub (OGH). The demonstration will present case studies of land use and land cover classification, including: (a) accessing and visualising data using OGH STAC; (b) building data cubes; (c) generating composite indexes; (d) combining multi-source data; (e) controlling the quality of training samples; (f) creating models using machine learning and deep learning algorithms; (g) tuning deep learning models; (h) parallel image classification using GPUs and CPUs; (i) removing outliers by post-processing; (j) using best practices for accuracy assessment. The workshop will use the open source R package SITS, which is one of the technologies whose development is being supported by the OEMC project.
Open-Earth-Monitor Cyberinfrastructure (Grant agreement ID: 101059548)
Please provide URL that you plan to use to distribute your materials (if available). –
Gilberto Câmara is a Brazilian researcher in Geoinformatics, GIScience, Spatial Analysis, and Land Use Modelling from INPE (Brazil's National Institute for Space Research). He was INPE’s director general (2005-2012), visiting professor at the Institute for Geoinformatics at the University of Münster (2013-2015), and Director of the Secretariat of GEO (Group on Earth Observations) from 2018-2021. He published 180 papers and cited more than 17,500 times with an H-index of 55 (Google Scholar, February 2024).
Rolf obtained his doctoral degree in Environmental Sciences in 2021 and has a background in geoinformatics, complex systems modeling, and business administration. At OpenGeoHub, he is a postdoctoral researcher.