Automatic seagrass banquettes detection from surveillance camera images with Detectron2

Authors

  • Gaetano Sabato University of Bari Author
  • Giovanni Scardino University of Bari Author
  • Alok Kushabaha University of Bari-IUSS Author
  • Marco Chirivì CETMA Author
  • Antonio Luparelli CETMA Author
  • Giovanni Scicchitano University of Bari Author

DOI:

https://doi.org/10.4461/GFDQ.2022.45.11

Keywords:

deep learning, seagrass, detection, beach monitoring

Abstract

In recent years, machine learning and deep learning methodologies have gained increasing attention in various fields of research, including environmental studies. Some algorithms with deep learning can be used to identify coastal features, detect changes over time, and monitor human activities on the coast, providing important information for sustainable coastal management. This study presents the application of the Detectron2 algorithm for monitoring a beach and verifying the presence or absence of stranded seagrass banquettes from video surveillance system images. The algorithm enables quick and automatic detection of these features, providing a valuable tool for beach managers and researchers alike. 

Published

2024-03-12

Issue

Section

Research and review papers

Similar Articles

You may also start an advanced similarity search for this article.