Unsupervised segmentation of underwater objects in sonar images

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We focus on the segmentation of sonar images for the aim of underwater object detection. Speckle noise and intensity inhomogeneity may cause false detections, and complex seabed textures like sand-ripples and sea-grass often lead to false segmentation. To tackle these problems, we propose a new method to incorporate the possible spatial correlation between neighboring pixels in the sonar image for improved segmentation. Our method modifies the expectation-maximization (EM) algorithm by adding an intermediate step (I-step) between the expectation (E-step) and maximization (M-step). Results show that our proposed method achieves improved segmentation performance over the state-of-the-art and is also robust to different seabed texture and for intensity inhomogeneity.

Original languageEnglish
Title of host publicationOCEANS 2017 - Aberdeen
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-5
Number of pages5
ISBN (Electronic)9781509052783
DOIs
StatePublished - 25 Oct 2017
EventOCEANS 2017 - Aberdeen - Aberdeen, United Kingdom
Duration: 19 Jun 201722 Jun 2017

Publication series

NameOCEANS 2017 - Aberdeen
Volume2017-October

Conference

ConferenceOCEANS 2017 - Aberdeen
Country/TerritoryUnited Kingdom
CityAberdeen
Period19/06/1722/06/17

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

Keywords

  • Expectation-maximization (EM)
  • gamma distribution
  • object detection
  • sand ripples
  • sonar image segmentation
  • speckle noise

ASJC Scopus subject areas

  • Instrumentation
  • Computer Networks and Communications
  • Oceanography
  • Acoustics and Ultrasonics
  • Automotive Engineering

Fingerprint

Dive into the research topics of 'Unsupervised segmentation of underwater objects in sonar images'. Together they form a unique fingerprint.

Cite this