Automatic Shoreline Detection from Eight-Band VHR Satellite Imagery

Abstract

Coastal erosion, which is naturally present in many areas of the world, can be significantly increased by factors such as the reduced transport of sediments as a result of hydraulic works carried out to minimize flooding. Erosion has a significant impact on both marine ecosystems and human activities; for this reason, several international projects have been developed to study monitoring techniques and propose operational methodologies. The increasing number of available high-resolution satellite platforms (i.e., Copernicus Sentinel) and algorithms to treat them allows the study of original approaches for the monitoring of the land in general and for the study of the coastline in particular. The present project aims to define a methodology for identifying the instantaneous shoreline, through images acquired from the WorldView 2 satellite, on eight spectral bands, with a geometric resolution of 0.5 m for the panchromatic image and 1.8 m for the multispectral one. A pixel-based classification methodology is used to identify the various types of land cover and to make combinations between the eight available bands. The experiments were carried out on a coastal area with contrasting morphologies. The eight bands in which the images are taken produce good results both in the classification process and in the combination of the bands, through the algorithms of normalized difference vegetation index (NDVI), normalized difference water index (NDWI), spectral angle mapper (SAM), and matched filtering (MF), with regard to the identification of the various soil coverings and, in particular, the separation line between dry and wet sand. In addition, the real applicability of an algorithm that extracts bathymetry in shallow water using the "coastal blue" band was tested. These data refer to the instantaneous shoreline and could be corrected in the future with morphological and tidal data of the coastal areas under study. Author Contributions: Supervision, M.A., V.B., R.B. and F.R.; Methodology, M.A., V.B., R.B. and F.R.; Investigation, M.A., V.B., R.B. and F.R.; Writing-Review and editing, V.B., R.B. and F.R. Funding: This research received no external funding. Conflicts of Interest:The authors declare no conflict of interest.

Remote Sensing-Based Automatic Detection of Shoreline Position: A Case Study in Apulia Region

“…Those promise to compensate for the limit imposed by the currently implemented satellite's temporal and spatial resolution providing data at a daily temporal resolution and at a higher spatial resolution. The capability of those satellites to retrieve the shoreline position has already been proved [91][92][93].…”

From electronic navigational chart data to sea-bottom models: Kriging approaches for the Bay of Pozzuoli

“…SBS is a less expansive system that MBS but provides much lower spatial resolution [9]. A good level of information about seabed morphology can be extract by multispectral satellite images, even if only in shallow water (depths less than 20 meters) [10], [11].…”

Detection of coastal dynamics depending on the Caspian Sea level change using remote sensing data: A case study in the Kura geomorphological sub-region (Azerbaijan)

“…Being located at the physical contact of land and water basin (Fu et al, 2017;Alicandro et al, 2019;Pugliano et al, 2019;Liu, 2017;Appeaning et al, 2008), coastal zones are natural territories with a complex geological-geomorphological structure, unique ecological environment, biological diversity and climatic conditions (Ashis, 2002;Zollini et al, 2020). In recent times, coastal areas have become vulnerable to the effects of natural disasters (typhoons, tornadoes, tsunamis, abrasions, floods, etc.)…”