A REVIEW AND TEST OF SHORELINE EXTRACTION TECHNIQUES

Abstract. Shoreline represents the boundary between land and sea, its accurate extraction is of utmost importance because economic ecological value coastal areas. Nowadays, satellite remote sensing widely used for monitoring natural environment. Indeed, data are cost-effective periodically available over large areas at a relatively high spatial resolution. Hence, automatic shoreline from images fundamental task management. methods usually applied to data. The goal this study compare performance different methods, such as thresholding more complex classification approaches, Random Forest (RF), Minimum Distance (MD), Maximum Likelihood (ML) K-means, using both optical radar images. considered case area shallow basin Orbetello Lagoon one ayre called Feniglia. supplier Copernicus program, which, through Sentinel-1 Sentinel-2 missions, provides medium-resolution, open-access products. accuracy obtained results methodologies checked by validating extracted an aerial orthomosaic and, subsequently, manually shoreline. A preliminary assessment was performed image classification, focusing on extracting four classes: water, soil, urban, forest, manual segmentation reference. In terms deviation reference shoreline, analysed achieved 3.75 m, less than half pixel size