Presentation 2021-11-12
Detecting moving ships in ALOS-2 spotlight SAR images by deep learning model
Takero Yoshida,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Spotlight SAR image has high resolution with long azimuth integration time. Meanwhile moving objects like ships are not focused due to the long integration time. Therefore, the moving ships in the spotlight SAR images are emerged as extended shapes in the azimuth direction. It is a typical imaging of moving ships in a spotlight mode, and therefore we can classify ships into moving and stationary cases. This paper shows the detection of moving ships by deep learning method, in particular using YOLOv5 model, which is relatively new method in a framework of YOLO family. ALOS-2/PALSAR-2 spotlight SAR images were analyzed and moving ships were detected by YOLOv5 model.
Keyword(in Japanese) (See Japanese page)
Keyword(in English)
Paper # SANE2021-57
Date of Issue 2021-11-04 (SANE)

Conference Information
Committee SANE
Conference Date 2021/11/11(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Online
Topics (in Japanese) (See Japanese page)
Topics (in English) ICSANE2021/Workshop on subsurface electromagnetic measurement
Chair Toshifumi Moriyama(Nagasaki Univ.)
Vice Chair Makoto Tanaka(Tokai Univ.) / Takeshi Amishima(Mitsubishi Electric)
Secretary Makoto Tanaka(Univ. of Tokyo) / Takeshi Amishima(ENRI)
Assistant Takayuki Kitamura(Mitsubishi Electric)

Paper Information
Registration To Technical Committee on Space, Aeronautical and Navigational Electronics
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Detecting moving ships in ALOS-2 spotlight SAR images by deep learning model
Sub Title (in English)
Keyword(1)
Keyword(2)
Keyword(3)
1st Author's Name Takero Yoshida
1st Author's Affiliation Tokyo University of Marine Science and Technology(TUMSAT)
Date 2021-11-12
Paper # SANE2021-57
Volume (vol) vol.121
Number (no) SANE-236
Page pp.pp.136-138(SANE),
#Pages 3
Date of Issue 2021-11-04 (SANE)