2025-09-02, 13:30–15:00 (Europe/Amsterdam), Expert Room 3
This tutorial introduces sits, an R package for Land use and land cover (LULC) classification using satellite image time series.
LULC classification are essential for environmental monitoring, agriculture, and climate change analysis. LULC maps provide key data for decision-making. sits offers an integrated framework to generate data cubes and classify satellite image time series. It provides an end-to-end solution integrating data management, machine learning, and validation.
The course covers the full LULC classification workflow. It starts with building data cubes and organizing satellite images in a structured way. Then, it moves to extracting and preparing time series samples. Next, participants train machine learning models to recognize LULC patterns. After that, they classify new satellite data using trained models. Finally, they validate results by assessing classification accuracy.
This course is for remote sensing experts, GIS professionals, and students interested in using machine learning for LULC analysis. It focuses on practical applications and offers hands-on experience with the entire process.
I recently defended my PhD thesis on using machine learning methods to analyze geospatial data. I'm now a senior lecturer at Adam Mickiewicz University in Poznan (Poland). I'm interested in remote sensing (especially in agriculture) and programming in R.