Ship-Iceberg Discrimination in Sentinel-2 Multispectral Imagery by Supervised Classification

Peder Heiselberg, Henning Heiselberg

Research output: Contribution to journalJournal articleResearchpeer-review

181 Downloads (Pure)

Abstract

The European Space Agency Sentinel-2 satellites provide multispectral images with pixel sizes down to 10 m. This high resolution allows for fast and frequent detection, classification and discrimination of various objects in the sea, which is relevant in general and specifically for the vast Arctic environment. We analyze several sets of multispectral image data from Denmark and Greenland fall and winter, and describe a supervised search and classification algorithm based on physical parameters that successfully finds and classifies all objects in the sea with reflectance above a threshold. It discriminates between objects like ships, islands, wakes, and icebergs, ice floes, and clouds with accuracy better than 90%. Pan-sharpening the infrared bands leads to classification and discrimination of ice floes and clouds better than 95%. For complex images with abundant ice floes or clouds, however, the false alarm rate dominates for small non-sailing boats.
Original languageEnglish
Article number1156
JournalRemote Sensing
Volume9
Issue number11
Number of pages19
ISSN2072-4292
DOIs
Publication statusPublished - 2017

Keywords

  • Sentinel-2
  • Multispectral
  • Ship
  • Iceberg
  • Detection
  • Discrimination
  • Classification
  • Arctic

Cite this

@article{3f8b76b9320a442faa1e380f4125450e,
title = "Ship-Iceberg Discrimination in Sentinel-2 Multispectral Imagery by Supervised Classification",
abstract = "The European Space Agency Sentinel-2 satellites provide multispectral images with pixel sizes down to 10 m. This high resolution allows for fast and frequent detection, classification and discrimination of various objects in the sea, which is relevant in general and specifically for the vast Arctic environment. We analyze several sets of multispectral image data from Denmark and Greenland fall and winter, and describe a supervised search and classification algorithm based on physical parameters that successfully finds and classifies all objects in the sea with reflectance above a threshold. It discriminates between objects like ships, islands, wakes, and icebergs, ice floes, and clouds with accuracy better than 90{\%}. Pan-sharpening the infrared bands leads to classification and discrimination of ice floes and clouds better than 95{\%}. For complex images with abundant ice floes or clouds, however, the false alarm rate dominates for small non-sailing boats.",
keywords = "Sentinel-2, Multispectral, Ship, Iceberg, Detection, Discrimination, Classification, Arctic",
author = "Peder Heiselberg and Henning Heiselberg",
year = "2017",
doi = "10.3390/rs9111156",
language = "English",
volume = "9",
journal = "Remote Sensing",
issn = "2072-4292",
publisher = "M D P I AG",
number = "11",

}

Ship-Iceberg Discrimination in Sentinel-2 Multispectral Imagery by Supervised Classification. / Heiselberg, Peder; Heiselberg, Henning.

In: Remote Sensing, Vol. 9, No. 11, 1156, 2017.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Ship-Iceberg Discrimination in Sentinel-2 Multispectral Imagery by Supervised Classification

AU - Heiselberg, Peder

AU - Heiselberg, Henning

PY - 2017

Y1 - 2017

N2 - The European Space Agency Sentinel-2 satellites provide multispectral images with pixel sizes down to 10 m. This high resolution allows for fast and frequent detection, classification and discrimination of various objects in the sea, which is relevant in general and specifically for the vast Arctic environment. We analyze several sets of multispectral image data from Denmark and Greenland fall and winter, and describe a supervised search and classification algorithm based on physical parameters that successfully finds and classifies all objects in the sea with reflectance above a threshold. It discriminates between objects like ships, islands, wakes, and icebergs, ice floes, and clouds with accuracy better than 90%. Pan-sharpening the infrared bands leads to classification and discrimination of ice floes and clouds better than 95%. For complex images with abundant ice floes or clouds, however, the false alarm rate dominates for small non-sailing boats.

AB - The European Space Agency Sentinel-2 satellites provide multispectral images with pixel sizes down to 10 m. This high resolution allows for fast and frequent detection, classification and discrimination of various objects in the sea, which is relevant in general and specifically for the vast Arctic environment. We analyze several sets of multispectral image data from Denmark and Greenland fall and winter, and describe a supervised search and classification algorithm based on physical parameters that successfully finds and classifies all objects in the sea with reflectance above a threshold. It discriminates between objects like ships, islands, wakes, and icebergs, ice floes, and clouds with accuracy better than 90%. Pan-sharpening the infrared bands leads to classification and discrimination of ice floes and clouds better than 95%. For complex images with abundant ice floes or clouds, however, the false alarm rate dominates for small non-sailing boats.

KW - Sentinel-2

KW - Multispectral

KW - Ship

KW - Iceberg

KW - Detection

KW - Discrimination

KW - Classification

KW - Arctic

U2 - 10.3390/rs9111156

DO - 10.3390/rs9111156

M3 - Journal article

VL - 9

JO - Remote Sensing

JF - Remote Sensing

SN - 2072-4292

IS - 11

M1 - 1156

ER -