Land cover fraction mapping across global biomes with Landsat data, spatially generalized regression models and spectral-temporal metrics

Mapping land cover in highly heterogeneous landscapes is challenging, and classifications have inherent limitations where the spatial resolution of remotely sensed data exceeds the size of small objects. For example, classifications based on 30-m Landsat data do not capture urban or other heterogene...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Remote sensing of environment Jg. 311; S. 114260
Hauptverfasser:	Schug, Franz, Pfoch, Kira A., Pham, Vu-Dong, van der Linden, Sebastian, Okujeni, Akpona, Frantz, David, Radeloff, Volker C.
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	Elsevier Inc 01.09.2024
Schlagworte:	ecosystems environment land cover Landsat Model transfer Neural network regression remote sensing Synthetic training data Weighted training data Synthetic training data Model transfer Neural network regression Weighted training data
ISSN:	0034-4257
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!