Course Description

This course provides an in-depth introduction to optical and hyperspectral remote sensing, emphasizing open-source and cloud-based workflows for environmental and agricultural monitoring. Participants explore the full analytical chain—from data acquisition and atmospheric correction to machine learning-based classification and biophysical parameter estimation. The course integrates theoretical lectures with guided exercises using real satellite datasets.

Learning Objectives

Upon completion, participants will be able to:

• Explain key principles of multispectral and hyperspectral remote sensing.
• Perform atmospheric correction using Sen2Cor and QGIS tools.
• Develop and utilize spectral libraries for material and vegetation analysis.
• Apply random forest classification and transfer learning techniques in crop monitoring.
• Implement drought and soil health indices using optical and hyperspectral data.
• Operate GRASS GIS and Google Earth Engine for integrated data analysis.

Duration: 8h

Topics:

(1) Optical Remote Sensing Fundamentals, covering multispectral imaging principles and atmospheric correction workflows using QGIS and Sen2Cor, spectral reflectance characteristics;

(2) Hyperspectral Data Analysis, (exploring legacy data EO-1/Hyperion, CHRIS/PROBA, and PRISMA and DESIS missions, spectral library development, and derivative spectroscopy);

(3) Machine Learning in Crop Monitoring, (implementing random forest classification and transfer learning);

(4) Drought & Soil Health Monitoring, (Modified Temperature Vegetation Dryness Index (MTVDI) and hyperspectral soil organic carbon (SOC) estimation using Google Earth Engine and GRASS GIS).

Technology

Throughout the course, participants gain hands-on experience with a diverse FOSS toolchain, including Sen2Cor, ENVI FOSS, QGIS, GRASS GIS, and Google Earth Engine.