Efficient Generation of Artificial Training DB for Ship Detection Using Satellite SAR Images

In this article, we propose an effective scheme to generate an artificial training database (DB) to mitigate the deficiency in the amount of training DB for ship detection using satellite synthetic aperture radar (SAR) images. In the proposed scheme, SAR signatures of ship targets are first obtained...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Seung-Jae Lee, Kwang-Jae Lee
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Electromagnetic numerical analysis (EMNA) satellite synthetic aperture radar ship detection synthetic aperture radar (SAR) remote sensing Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
Acceso en línea:	https://doaj.org/article/318a29962db044c38e41e68128fccc69
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:318a29962db044c38e41e68128fccc69
record_format	dspace
spelling	oai:doaj.org-article:318a29962db044c38e41e68128fccc692021-12-02T00:00:05ZEfficient Generation of Artificial Training DB for Ship Detection Using Satellite SAR Images2151-153510.1109/JSTARS.2021.3128184https://doaj.org/article/318a29962db044c38e41e68128fccc692021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9615033/https://doaj.org/toc/2151-1535In this article, we propose an effective scheme to generate an artificial training database (DB) to mitigate the deficiency in the amount of training DB for ship detection using satellite synthetic aperture radar (SAR) images. In the proposed scheme, SAR signatures of ship targets are first obtained from two different types of image: 1) Korea Multipurpose Satellite-5 images with automatic identification system information and 2) simulated SAR images that are acquired using electromagnetic numerical analysis. Then, the ship signatures are combined with sea clutter models to generate realistic SAR patches. For overall ship detection tasks (i.e., training and testing), the oriented bounding box concept was adopted to appropriately represent slender ship targets in SAR images. When compared with the rash collection of a large number of real SAR images, the proposed scheme is more efficient in terms of cost and time. Experimental results demonstrated that the proposed method can considerably aid in increasing the ship detection capacity of deep learning models. Thus, it is expected that the proposed scheme can be usefully exploited to solve the lack of the amount of training DB.Seung-Jae LeeKwang-Jae LeeIEEEarticleElectromagnetic numerical analysis (EMNA)satellite synthetic aperture radarship detectionsynthetic aperture radar (SAR) remote sensingOcean engineeringTC1501-1800Geophysics. Cosmic physicsQC801-809ENIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 14, Pp 11764-11774 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Electromagnetic numerical analysis (EMNA) satellite synthetic aperture radar ship detection synthetic aperture radar (SAR) remote sensing Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
spellingShingle	Electromagnetic numerical analysis (EMNA) satellite synthetic aperture radar ship detection synthetic aperture radar (SAR) remote sensing Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809 Seung-Jae Lee Kwang-Jae Lee Efficient Generation of Artificial Training DB for Ship Detection Using Satellite SAR Images
description	In this article, we propose an effective scheme to generate an artificial training database (DB) to mitigate the deficiency in the amount of training DB for ship detection using satellite synthetic aperture radar (SAR) images. In the proposed scheme, SAR signatures of ship targets are first obtained from two different types of image: 1) Korea Multipurpose Satellite-5 images with automatic identification system information and 2) simulated SAR images that are acquired using electromagnetic numerical analysis. Then, the ship signatures are combined with sea clutter models to generate realistic SAR patches. For overall ship detection tasks (i.e., training and testing), the oriented bounding box concept was adopted to appropriately represent slender ship targets in SAR images. When compared with the rash collection of a large number of real SAR images, the proposed scheme is more efficient in terms of cost and time. Experimental results demonstrated that the proposed method can considerably aid in increasing the ship detection capacity of deep learning models. Thus, it is expected that the proposed scheme can be usefully exploited to solve the lack of the amount of training DB.
format	article
author	Seung-Jae Lee Kwang-Jae Lee
author_facet	Seung-Jae Lee Kwang-Jae Lee
author_sort	Seung-Jae Lee
title	Efficient Generation of Artificial Training DB for Ship Detection Using Satellite SAR Images
title_short	Efficient Generation of Artificial Training DB for Ship Detection Using Satellite SAR Images
title_full	Efficient Generation of Artificial Training DB for Ship Detection Using Satellite SAR Images
title_fullStr	Efficient Generation of Artificial Training DB for Ship Detection Using Satellite SAR Images
title_full_unstemmed	Efficient Generation of Artificial Training DB for Ship Detection Using Satellite SAR Images
title_sort	efficient generation of artificial training db for ship detection using satellite sar images
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/318a29962db044c38e41e68128fccc69
work_keys_str_mv	AT seungjaelee efficientgenerationofartificialtrainingdbforshipdetectionusingsatellitesarimages AT kwangjaelee efficientgenerationofartificialtrainingdbforshipdetectionusingsatellitesarimages
_version_	1718403975303135232

Efficient Generation of Artificial Training DB for Ship Detection Using Satellite SAR Images

Ejemplares similares