1 Department of Design og Kommunikation, Faculty of Humanities, SDU2 Faculty of Humanities, SDU3 Centre for Tourism, Innovation & Culture, Department of Design og Kommunikation, Faculty of Humanities, SDU4 Centre for Tourism, Innovation and Culture, Department of History, Faculty of Humanities, SDU5 Lancaster University6 Centre for Tourism, Innovation and Culture, Department of History, Faculty of Humanities, SDU
In this paper, the moving-windows approach to calculation and analysis of spatial metrics is tested with particular focus on forest mapping. The influence of window size on average metrics values, agreement between values from different EO-based data sources and local variance of metrics values is analysed using standard statistical approaches. Forest Concentration Profiles, based on forest-non forest masks for moving windows is presented as an approach to characterise structure and distribution of forest over certain areas of interest.
Remote Sensing and Geographical Informations Systems for Environmental Studies, Applications in Forestry: Schriften Aus Der Forstlichen Fakultät Der Universität Göttingen Und Der Niedersächsischen Forstlichen Versuchsanstalt, 2005, p. 184-194