Machine learning algorithms of remote sensing data processing for mapping changes in land cover types over Central Apennines, Italy

These slides present research on mapping land cover types in Central Apennines, Italy. Monitoring biodiversity of the Mediterranean region is fundamental for land management in protected areas. Mountainous landscapes of Central Italy have large variations of patches and complex mosaic of vegetation. The machine learning (ML) algorithms of remote sensing (RS) data processing were applied to solve the challenges of heterogeneous patterns recognition. The scripting cartographic software GRASS GIS was used for satellite image processing Landsat 8-9 OLI/TIRS. Several algorithms of ML methods were tested, compared and used for image analysis. The results present a time series analysis of several satellite images on period of 2014-2024. The slides were presented at the workshop "A data-driven approach to reconstruct vegetation change in rewilding areas" organised by BIOME – Biodiversity & MacroEcology Laboratory, Dipartimento di Scienze Biologiche, Geologiche ed Ambientali – BiGeA Alma Mater Studiorum - University of Bologna on 11-13 February 2025 in Bologna, Italy.