EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Water Body Automatic Detection in Remotely Sensed Images: A New Index and Methodological Approach for Sentinel-2 Satellite Datasets

Francesca Guastaferro, Pasquale Maglione, Claudio Parente, Andrea Vallario

Remote Sensing · 06.11.2026 · 18 (21) · 3821

DOI: 10.3390/rs18213821 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 06.11.2026. Verfasst von Francesca Guastaferro und 3 weiteren Autor:innen. Die Arbeit nennt 117 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: PRIN 2022 PNRR.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This article proposes a fully automated engineering pipeline for extracting surface water from Sentinel-2 satellite imagery using a novel spectral index and unsupervised three-cluster classification. The proposed Water Body Enhancement Index (WBEI) relies exclusively on native 10 m spatial resolution bands (Blue, Green, Red, and Near-Infrared [NIR]), thereby eliminating the geometric degradation caused by Short-Wave Infrared (SWIR) band resampling. To ensure methodological robustness across diverse global aquatic environments, WBEI is validated across six distinct experimental sites located in Italy, Australia, the USA, and Mozambique, utilizing target test fields of no less than 100,000 pixels per scene. Performance is rigorously compared via an identical automated framework against widely used conventional water indices: Automated Water Extraction Index-Shadow version (AWEIsh), Modified Normalized Difference Water Index (MNDWI), Normalized Difference Water Index (NDWI), and Water Ratio Index (WRI). Validation metrics include traditional confusion matrices (overall, producer’s, and user’s accuracy) and the spatial Ratio Index (RI) expressed in meters to precisely evaluate shoreline displacement. Experimental results indicate that the proposed WBEI consistently achieves superior extraction accuracy, with overall accuracy values exceeding 94.9% in all geographic cases, offering a highly reliable, reproducible, and fully automated pipeline managed via a Geographic Information System (GIS).

Förderung

  • PRIN 2022 PNRR

Zitieren

APA

Guastaferro, F., Maglione, P., Parente, C., & Vallario, A. (2026). Water Body Automatic Detection in Remotely Sensed Images: A New Index and Methodological Approach for Sentinel-2 Satellite Datasets. Remote Sensing, 18(21), 3821. https://doi.org/10.3390/rs18213821

Vancouver

Guastaferro F, Maglione P, Parente C, Vallario A. Water Body Automatic Detection in Remotely Sensed Images: A New Index and Methodological Approach for Sentinel-2 Satellite Datasets. Remote Sensing. 2026;18(21):3821. doi:10.3390/rs18213821

BibTeX
@article{Guastaferro2026,
  author = {Guastaferro, Francesca and Maglione, Pasquale and Parente, Claudio and Vallario, Andrea},
  title = {Water Body Automatic Detection in Remotely Sensed Images: A New Index and Methodological Approach for Sentinel-2 Satellite Datasets},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {21},
  pages = {3821},
  doi = {10.3390/rs18213821}
}
Kostenlos starten →