First Results on Wake Detection in SAR Images by Deep Learning

Spaceborne synthetic aperture radar (SAR) represents a powerful source of data for enhancing maritime domain awareness (MDA). Wakes generated by traveling vessels hold a crucial role in MDA since they can be exploited both for ship route and velocity estimation and as a marker of ship presence. Even...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Roberto Del Prete, Maria Daniela Graziano, Alfredo Renga
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	automatic target recognition deep learning ship wake detection synthetic aperture radar Science Q
Acceso en línea:	https://doaj.org/article/4115cbd9e98440b0ae9aa430778c4cdd
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:4115cbd9e98440b0ae9aa430778c4cdd
record_format	dspace
spelling	oai:doaj.org-article:4115cbd9e98440b0ae9aa430778c4cdd2021-11-25T18:54:28ZFirst Results on Wake Detection in SAR Images by Deep Learning10.3390/rs132245732072-4292https://doaj.org/article/4115cbd9e98440b0ae9aa430778c4cdd2021-11-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/22/4573https://doaj.org/toc/2072-4292Spaceborne synthetic aperture radar (SAR) represents a powerful source of data for enhancing maritime domain awareness (MDA). Wakes generated by traveling vessels hold a crucial role in MDA since they can be exploited both for ship route and velocity estimation and as a marker of ship presence. Even if deep learning (DL) has led to an impressive performance boost on a variety of computer vision tasks, its usage for automatic target recognition (ATR) in SAR images to support MDA is still limited to the detection of ships rather than ship wakes. A dataset is presented in this paper and several state-of-the-art object detectors based on convolutional neural networks (CNNs) are tested with different backbones. The dataset, including more than 250 wake chips, is realized by visually inspecting Sentinel-1 images over highly trafficked maritime sites. Extensive experiments are shown to characterize CNNs for the wake detection task. For the first time, a deep-learning approach is implemented to specifically detect ship wakes without any a-priori knowledge or cuing about the location of the vessel that generated the wake. No annotated dataset was available to train deep-learning detectors on this task, which is instead presented in this paper. Moreover, the benchmarks achieved for different detectors point out promising features and weak points of the relevant approaches. Thus, the work also aims at stimulating more research in this promising, but still under-investigated, field.Roberto Del PreteMaria Daniela GrazianoAlfredo RengaMDPI AGarticleautomatic target recognitiondeep learningship wake detectionsynthetic aperture radarScienceQENRemote Sensing, Vol 13, Iss 4573, p 4573 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	automatic target recognition deep learning ship wake detection synthetic aperture radar Science Q
spellingShingle	automatic target recognition deep learning ship wake detection synthetic aperture radar Science Q Roberto Del Prete Maria Daniela Graziano Alfredo Renga First Results on Wake Detection in SAR Images by Deep Learning
description	Spaceborne synthetic aperture radar (SAR) represents a powerful source of data for enhancing maritime domain awareness (MDA). Wakes generated by traveling vessels hold a crucial role in MDA since they can be exploited both for ship route and velocity estimation and as a marker of ship presence. Even if deep learning (DL) has led to an impressive performance boost on a variety of computer vision tasks, its usage for automatic target recognition (ATR) in SAR images to support MDA is still limited to the detection of ships rather than ship wakes. A dataset is presented in this paper and several state-of-the-art object detectors based on convolutional neural networks (CNNs) are tested with different backbones. The dataset, including more than 250 wake chips, is realized by visually inspecting Sentinel-1 images over highly trafficked maritime sites. Extensive experiments are shown to characterize CNNs for the wake detection task. For the first time, a deep-learning approach is implemented to specifically detect ship wakes without any a-priori knowledge or cuing about the location of the vessel that generated the wake. No annotated dataset was available to train deep-learning detectors on this task, which is instead presented in this paper. Moreover, the benchmarks achieved for different detectors point out promising features and weak points of the relevant approaches. Thus, the work also aims at stimulating more research in this promising, but still under-investigated, field.
format	article
author	Roberto Del Prete Maria Daniela Graziano Alfredo Renga
author_facet	Roberto Del Prete Maria Daniela Graziano Alfredo Renga
author_sort	Roberto Del Prete
title	First Results on Wake Detection in SAR Images by Deep Learning
title_short	First Results on Wake Detection in SAR Images by Deep Learning
title_full	First Results on Wake Detection in SAR Images by Deep Learning
title_fullStr	First Results on Wake Detection in SAR Images by Deep Learning
title_full_unstemmed	First Results on Wake Detection in SAR Images by Deep Learning
title_sort	first results on wake detection in sar images by deep learning
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/4115cbd9e98440b0ae9aa430778c4cdd
work_keys_str_mv	AT robertodelprete firstresultsonwakedetectioninsarimagesbydeeplearning AT mariadanielagraziano firstresultsonwakedetectioninsarimagesbydeeplearning AT alfredorenga firstresultsonwakedetectioninsarimagesbydeeplearning
_version_	1718410581938012160

First Results on Wake Detection in SAR Images by Deep Learning

Ejemplares similares