Andere Kunden interessierten sich auch für
![Remote Sensing in Forest Fire Monitoring and Post-fire Damage Analysis]( "Remote Sensing in Forest Fire Monitoring and Post-fire Damage Analysis")
Remote Sensing in Forest Fire Monitoring and Post-fire Damage Analysis92,99 €![Advances in Remote Sensing for Global Forest Monitoring]( "Advances in Remote Sensing for Global Forest Monitoring")
Advances in Remote Sensing for Global Forest Monitoring105,99 €![Water Quality Modelling Using Statistical Analysis and Remote Sensing]( "Water Quality Modelling Using Statistical Analysis and Remote Sensing")
Water Quality Modelling Using Statistical Analysis and Remote Sensing61,99 €![Crops and Vegetation Monitoring with Remote/Proximal Sensing]( "Crops and Vegetation Monitoring with Remote/Proximal Sensing")
Crops and Vegetation Monitoring with Remote/Proximal Sensing104,99 €![Spatio-Temporal Analysis of Urbanization Using GIS and Remote Sensing in Developing Countries]( "Spatio-Temporal Analysis of Urbanization Using GIS and Remote Sensing in Developing Countries")
Spatio-Temporal Analysis of Urbanization Using GIS and Remote Sensing in Developing Countries95,99 €![Advanced Machine Learning and Deep Learning Approaches for Remote Sensing II]( "Advanced Machine Learning and Deep Learning Approaches for Remote Sensing II")
Advanced Machine Learning and Deep Learning Approaches for Remote Sensing II117,99 €![Upwelling and fish identification by remote sensing]( "Upwelling and fish identification by remote sensing")
Upwelling and fish identification by remote sensing44,99 €
Forest structure, e.g. the composition and distribution of tree species, is a key element for characterizing ecological functions and ecological state of forest ecosystems. The aim of this master thesis was to model 15 forest structure measures of a mixed temperate forest in the Bavarian Forest National Park (Germany) with hyperspectral remote sensing data (HyMap). The findings indicated that hyperspectral data has high potential to identify forest structure even in heterogeneous mixed forests like the Bavarian Forest National Park. Forest structure measures were derived from vegetation surveys on 102 ground plots and served as dependent variables in a decision tree based random forest model. The independent variables were obtained from hyperspectral data in 7 m resolution, which was transformed by a minimum noise fraction rotation (MNF). With these two datasets random forest model performance on each forest structure measure was compared to model performance derived from literature. Furthermore, descriptive statistics, correlation analysis and ordination methods were used to discuss the results.