Oil spills, caused by accidents or by ships cleaning their tanks, represent big threats for maritime and coastal ecosystems health. A very effective detection of oil spills can be performed using satellite synthetic aperture radar (SAR) systems, operating regardless of cloud coverage and sunlight and capable of discriminating oil from regular sea surface. However, discriminating between real oil spills and look-alikes (such as natural oils and seepages, often occurring in upwelling sea areas), although well performed by expert SAR image interpreters, poses a great challenge for automatic processes. In addition, a visual check performed by human operators on a great number of images would be too expensive. Therefore, many solutions for automatic detection have been tried in the last few years, using probabilistic models and, more recently, machine learning. This work presents an innovative solution based on image-to-image translation using convolutional neural networks (CNNs) trained with an adversarial loss function. The proposed approach has been tested, with very promising results, using Radarsat-2 and Sentinel-1 SAR data over the Mediterranean Sea and some areas of the Atlantic Ocean and the North Sea.

An adversarial learning approach for oil spill detection from SAR images / Ronci, F.; Avolio, C.; Di Donna, M.; Zavagli, M.; Piccialli, V.; Costantini, M.. - (2020). (Intervento presentato al convegno 2020 IEEE Radar Conference, RadarConf 2020 tenutosi a Firenze; Italia) [10.1109/RadarConf2043947.2020.9266475].

An adversarial learning approach for oil spill detection from SAR images

Piccialli V.;
2020

Abstract

Oil spills, caused by accidents or by ships cleaning their tanks, represent big threats for maritime and coastal ecosystems health. A very effective detection of oil spills can be performed using satellite synthetic aperture radar (SAR) systems, operating regardless of cloud coverage and sunlight and capable of discriminating oil from regular sea surface. However, discriminating between real oil spills and look-alikes (such as natural oils and seepages, often occurring in upwelling sea areas), although well performed by expert SAR image interpreters, poses a great challenge for automatic processes. In addition, a visual check performed by human operators on a great number of images would be too expensive. Therefore, many solutions for automatic detection have been tried in the last few years, using probabilistic models and, more recently, machine learning. This work presents an innovative solution based on image-to-image translation using convolutional neural networks (CNNs) trained with an adversarial loss function. The proposed approach has been tested, with very promising results, using Radarsat-2 and Sentinel-1 SAR data over the Mediterranean Sea and some areas of the Atlantic Ocean and the North Sea.
2020
2020 IEEE Radar Conference, RadarConf 2020
Deep Learning; GAN; Oil Spill; SAR; Semantic Segmentation; U-Net
04 Pubblicazione in atti di convegno::04b Atto di convegno in volume
An adversarial learning approach for oil spill detection from SAR images / Ronci, F.; Avolio, C.; Di Donna, M.; Zavagli, M.; Piccialli, V.; Costantini, M.. - (2020). (Intervento presentato al convegno 2020 IEEE Radar Conference, RadarConf 2020 tenutosi a Firenze; Italia) [10.1109/RadarConf2043947.2020.9266475].
File allegati a questo prodotto
File Dimensione Formato  
Ronci_An-Adversarial-oil_2020.pdf

solo gestori archivio

Tipologia: Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 579.25 kB
Formato Adobe PDF
579.25 kB Adobe PDF   Contatta l'autore

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

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/1623070
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 5
  • ???jsp.display-item.citation.isi??? 4
social impact