| Popis: |
The cork oak forest of Benslimane province, which lies at its southern limit on a world scale, is characterized by the heterogeneity of its forest stands and the fragmentation of its forest area. These characteristics result from the interaction of various factors, including anthropogenic factors which are further aggravated by climate change. The heterogeneity of species that occupy the soil are represented by mixed pixels, which limits the reliability of conventional methods of classification. In the present study, we propose to apply a specific sub-pixel classification technique to spatial imagery from the Operational Land Imager (OLI) sensor of Landsat 8 to disaggregate the information and thus recover the spectral signatures specific to each \"pure\" component of the populations of the ecosystems studied. |
