Remote sensing

Implementing land cover/land use (LCLU) classification based on Earth observation data and ML/AI classifiers needs a substantial amount of LCLU samples covering the study area.

The goal of this nugget is to introduce you to different methods and techniques for detecting and analysing LCLU changes using EO time-series data in Google Earth Engine (GEE).

This module introduces students to citizen science data for Earth Observation, focusing on handling crowdsourced land cover and land use (LCLU) data samples in the R program. It covers the statistical accuracy assessment of citizen science data and the preparation of Collect Earth Online (CEO) LCLU samples data for satellite image classification.

Learning outcomes include understanding SAR remote sensing technology and using Sentinel 1 SAR images for built-up area mapping.

Classify multispectral satellite images available in different platforms and download and manage the data sets using different open-source tools. You will be able to carry out a basic multispectral image processing flow based on supervised ML classification methods.

Key steps of managing the data using the Sentinel Application Platform (SNAP) this includes installation of SNAP, use of different SNAP tools for performing basic visualization of the Sentinel 2 images as well as installation of a specific plugin for image processing.

Key concepts of environmental remote sensing and practical understanding of the importance of Earth observation in the identification, mapping, monitoring, analysis and modelling of our planet. After the completion of the tasks, you are able to argue for the significance of remote sensing in Earth observation, and you have gained practical skills in searching, exploring