“Machine Learning with Earth Observation data: Case studies with Semantic Segmentation and Regression” tutorial
An educational material on the use of machine learning methods applied to Earth Observation data. The material was prepared for researchers, educators and professionals in local, regional, national or international agencies with an intermediate knowledge of geospatial information and usage of QGIS. The following educational material has been produced within the framework of the 2021 OSGeo UN Committee Educational Challenge entitled “Challenge 1: Training on Satellite Data Analysis and Machine Learning with QGIS (Satellite_QGIS)”. This tutorial proposes a methodology for segmentation of snow areas in the Huascaran mountain, located in Peru, to monitor ice loss. The Semi-Automatic Classification Plugin (SCP) and the Dzetsaka plugin are employed for this purpose, to download and pre-process satellite data, and to train a machine learning algorithm for classification.
