Assessing the Prediction Accuracy of Geomorphon-Based Automated Landform Classification: An Example from the Ionian Coastal Belt of Southern Italy

Automatic procedures for landform extraction is a growing research field but extensive quantitative studies of the prediction accuracy Landform Classification (ACL) based on direct comparison with geomorphological maps are rather limited. In this work, we test an algorithm automatic classification large sector Ionian coast southern Italian belt through detailed map. was performed by using individuation basic classes named geomorphons. Spatial overlay between main mapped landforms deriving from traditional analysis and results highlighted satisfactory percentage (higher than 70%) geomorphon-based method coastal plain area drainage network. The decreased about 20–30% marine fluvial terraces, while overall ACL map 69%. Our suggest that could represent robust tool to recognize elements landscape at scale, which can be useful advanced steps mapping such as genetic interpretation delineation complex composite geomorphic elements.