Improved Automatic Classification of Litho-Geomorphological Units by Using Raster Image Blending, Vipava Valley (SW Slovenia)

Automatic landslide classification based on digital elevation models has become a powerful complementary tool to field mapping. Many studies focus the automatic of landslides’ geomorphological features, such as their steep main scarps, but in many cases, scarps and other morphological features are difficult for algorithms detect. In this study, we performed an different litho-geomorphological units differentiate slope mass movements maps by using Maximum Likelihood Classification. The was high-resolution lidar-derived DEM Vipava Valley, SW Slovenia. results show improvement over previous approaches used blended image (VAT, which included four raster layers with weights) along common morphometric analysis surface (e.g., slope, elevation, aspect, TRI, curvature, etc.). newly created map showed better five classes study recognizes alluvial deposits, carbonate cliffs (including scarps), plateaus, flysch, deposits than studies. Multivariate statistics recognized VAT layer most important highest eigenvalues, when combined Aspect Elevation layers, it explained 90% total variance. paper also discusses correlations between suited certain analyses.