Skip to main navigation Skip to search Skip to main content

Semiautomated Segmentation of Sentinel-1 SAR Imagery for Mapping Sea Ice in Labrador Coast

  • Weikai Tan
  • , Jonathan Li*
  • , Linlin Xu
  • , Michael A. Chapman
  • *Corresponding author for this work
  • University of Waterloo
  • Xiamen University
  • China University of Geosciences, Beijing
  • Toronto Metropolitan University

Research output: Contribution to journal › Article › peer-review

Abstract

This study aims at proposing a semiautomated sea ice segmentation workflow utilizing Sentinel-1 synthetic aperture radar imagery. The workflow consists of two main steps. First, preferable features in sea ice interpretation were determined with a random forest feature selection method. Second, an unsupervised graph-cut image segmentation was performed. The workflow was tested on 13 Sentinel-1A images from January to June 2016, and the results were evaluated on open water segmentation per ice charts provided by Canada Ice Service. The results showed that the proposed workflow was able to segment Sentinel-1 images in to appropriate number of classes, and the potential water identification rate reached 95%.

Original languageEnglish
Pages (from-to)1419-1432
Number of pages14
JournalIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Volume11
Issue number5
DOIs
StatePublished - May 2018
Externally publishedYes

Keywords

  • Image segmentation
  • sea ice
  • sentinel-1
  • synthetic aperture radar

Fingerprint

Dive into the research topics of 'Semiautomated Segmentation of Sentinel-1 SAR Imagery for Mapping Sea Ice in Labrador Coast'. Together they form a unique fingerprint.

Cite this