University of Tasmania
Browse

File(s) under permanent embargo

Ship detection in Sentinel-1 imagery using the H-dome transformation

conference contribution
posted on 2023-05-23, 10:22 authored by Schwegmann, CP, Kleynhans, W, Brian SalmonBrian Salmon, Mdakane, L
The advancement of ship detections is an important focus of Maritime Domain Awareness. Improvements in both speed and correctness responses to events can be made by introducing new ship detection methods. Sentinel-1 imagery has recently been released for public usage and with it the chance to test new methods on current data. By exploiting the fact that ships are locally bright objects, this paper presents a method for ship detection within Sentinel-1 imagery using the H-dome transform. This transformation, typically used to highlight sub-cellular structures in biology, can be used to flatten clutter around the ship and highlight the brightest section of ships. The method was tested against two Sentinel-1 images in both HH and HV polarization with a total of 82 ships. The method provided improved FAR over the conventional CA-CFAR method at the cost of slightly worse detection accuracies. Future work includes using the method as auxiliary data in other ship detection methods.

History

Publication title

2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS)

Editors

IEEE

Pagination

3711-3714

ISBN

978-1-4799-7929-5

Department/School

School of Engineering

Publisher

Institute of Electrical and Electronics Engineers

Place of publication

United States of America

Event title

International Geoscience and Remote Sensing Symposium 2015

Event Venue

Milan, Italy

Date of Event (Start Date)

2015-07-26

Date of Event (End Date)

2015-07-31

Rights statement

Copyright 2015 IEEE

Repository Status

  • Restricted

Socio-economic Objectives

Expanding knowledge in engineering

Usage metrics

    University Of Tasmania

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC