Marine litter is harmful to coastal and ocean environments for many reasons. Many devices have been developed in the last few years to recover as much litter as possible in ports and coastal areas. However, they usually employ a brute-force approach, leading to high energy and resource consumption. This is not negligible, considering their ecological goal. In this scenario, the deployment of Unmanned Surface Vehicles (USVs) to inspect the area of interest could complement the use of cleaning devices, adding relevant knowledge and intelligence to the context. Cleaning devices could be deployed or activated only where and when needed, leading to a more resource-aware approach. The contribution of this work is the development and deployment of an object detection neural network onto the H20mni-X USV, aiming to real-time detection of floating marine litter through an optical camera.

Real-Time Floating Marine Litter Detection on USV / Lazzerini, Guido; Topini, Alberto; Liverani, Gherardo; Cecchi, Lorenzo; Ridolfi, Alessandro; Ferreira, Fausto. - ELETTRONICO. - (2024), pp. 148-153. (Intervento presentato al convegno 2024 IEEE International Workshop on Metrology for the Sea, MetroSea 2024 tenutosi a Portorose, Slovenia nel 14-16 October 2024) [10.1109/metrosea62823.2024.10765677].

Real-Time Floating Marine Litter Detection on USV

Lazzerini, Guido
;
Topini, Alberto;Liverani, Gherardo;Cecchi, Lorenzo;Ridolfi, Alessandro;
2024

Abstract

Marine litter is harmful to coastal and ocean environments for many reasons. Many devices have been developed in the last few years to recover as much litter as possible in ports and coastal areas. However, they usually employ a brute-force approach, leading to high energy and resource consumption. This is not negligible, considering their ecological goal. In this scenario, the deployment of Unmanned Surface Vehicles (USVs) to inspect the area of interest could complement the use of cleaning devices, adding relevant knowledge and intelligence to the context. Cleaning devices could be deployed or activated only where and when needed, leading to a more resource-aware approach. The contribution of this work is the development and deployment of an object detection neural network onto the H20mni-X USV, aiming to real-time detection of floating marine litter through an optical camera.
2024
2024 IEEE International Workshop on Metrology for the Sea, MetroSea 2024 - Proceedings
2024 IEEE International Workshop on Metrology for the Sea, MetroSea 2024
Portorose, Slovenia
14-16 October 2024
Goal 9: Industry, Innovation, and Infrastructure
Lazzerini, Guido; Topini, Alberto; Liverani, Gherardo; Cecchi, Lorenzo; Ridolfi, Alessandro; Ferreira, Fausto
File in questo prodotto:
File Dimensione Formato  
Real-Time_Floating_Marine_Litter_Detection_on_USV.pdf

Accesso chiuso

Descrizione: Articolo principale
Tipologia: Pdf editoriale (Version of record)
Licenza: Tutti i diritti riservati
Dimensione 3.8 MB
Formato Adobe PDF
3.8 MB Adobe PDF   Richiedi una copia

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/1410475
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact