Multispectral Sentinel-2 and SAR Sentinel-1 Integration for Automatic Land Cover Classification

The study of land cover and use dynamics are fundamental to understanding the radical changes that human activity is causing locally globally analyse continuous metamorphosis landscape. In Europe, Copernicus Program offers numerous territorial monitoring tools users decision makers, such as Sentinel data. This research aims at developing implementing a mapping change detection methodology through classification Sentinel-1 Sentinel-2 satellite goal create versatile economically sustainable algorithm capable rapidly processing large amounts data, allowing creation national-scale products with high spatial resolution update frequency for operational purposes. Great attention was paid compatibility main activities planned in near future national European level. this sense, system consistent specifications EAGLE group has been adopted. involves definition distinct sets rules each macro-classes classes. refers pixels’ spectral backscatter characteristics, exploiting multi-temporal indices while proposing two new ones: NDCI distinguish between broad-leaved needle-leaved trees, Burned Index (BI) identify burned areas. allowed production map 2018 related forest disturbances consumption 2017–2018, reaching an overall accuracy 83%.