Utilizing Sentinel-1A Radar Images for Large-Area Land Cover Mapping with Machine-learning Methods

Mohammad Imangholiloo (Corresponding Author), Jussi Rasinmäki, Yrjö Rauste, Markus Holopainen

Research output: Contribution to journalArticleScientificpeer-review

Abstract

Land use and land cover maps are vital sources of information for many applications. Recently, using high-resolution and open-access satellite images has become a preferred method for mapping land cover, especially over large areas. This study was designed to map the land cover and agricultural fields of a large area using Sentinel-1A synthetic aperture radar (SAR) images. Seven machine-learning methods were employed for image analyses. The Random Forest classifier algorithm outperformed the other machine-learning methods in the training step; thus, we selected it for further use and tuned its parameters. After several image processing steps, we classified the final image into 23 land cover classes and achieved an overall accuracy of 42% for all classes, and 57% for agricultural fields. This research note highlights some characteristics and advantages of using Sentinel-1A images and provides novel methods for nation-wide large-area mapping applications.

Original languageEnglish
Pages (from-to)163-175
Number of pages13
JournalCanadian Journal of Remote Sensing
Volume45
Issue number2
DOIs
Publication statusPublished - 2019
MoE publication typeA1 Journal article-refereed

Fingerprint

land cover
radar
image processing
synthetic aperture radar
land use
machine learning
method

Keywords

  • radar
  • land cover

Cite this

Imangholiloo, Mohammad ; Rasinmäki, Jussi ; Rauste, Yrjö ; Holopainen, Markus. / Utilizing Sentinel-1A Radar Images for Large-Area Land Cover Mapping with Machine-learning Methods. In: Canadian Journal of Remote Sensing. 2019 ; Vol. 45, No. 2. pp. 163-175.
@article{ffb4ac7f6ea2491dbe29f604e1379dc1,
title = "Utilizing Sentinel-1A Radar Images for Large-Area Land Cover Mapping with Machine-learning Methods",
abstract = "Land use and land cover maps are vital sources of information for many applications. Recently, using high-resolution and open-access satellite images has become a preferred method for mapping land cover, especially over large areas. This study was designed to map the land cover and agricultural fields of a large area using Sentinel-1A synthetic aperture radar (SAR) images. Seven machine-learning methods were employed for image analyses. The Random Forest classifier algorithm outperformed the other machine-learning methods in the training step; thus, we selected it for further use and tuned its parameters. After several image processing steps, we classified the final image into 23 land cover classes and achieved an overall accuracy of 42{\%} for all classes, and 57{\%} for agricultural fields. This research note highlights some characteristics and advantages of using Sentinel-1A images and provides novel methods for nation-wide large-area mapping applications.",
keywords = "radar, land cover",
author = "Mohammad Imangholiloo and Jussi Rasinm{\"a}ki and Yrj{\"o} Rauste and Markus Holopainen",
year = "2019",
doi = "10.1080/07038992.2019.1635877",
language = "English",
volume = "45",
pages = "163--175",
journal = "Canadian Journal of Remote Sensing",
issn = "0703-8992",
publisher = "Taylor & Francis",
number = "2",

}

Utilizing Sentinel-1A Radar Images for Large-Area Land Cover Mapping with Machine-learning Methods. / Imangholiloo, Mohammad (Corresponding Author); Rasinmäki, Jussi; Rauste, Yrjö; Holopainen, Markus.

In: Canadian Journal of Remote Sensing, Vol. 45, No. 2, 2019, p. 163-175.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Utilizing Sentinel-1A Radar Images for Large-Area Land Cover Mapping with Machine-learning Methods

AU - Imangholiloo, Mohammad

AU - Rasinmäki, Jussi

AU - Rauste, Yrjö

AU - Holopainen, Markus

PY - 2019

Y1 - 2019

N2 - Land use and land cover maps are vital sources of information for many applications. Recently, using high-resolution and open-access satellite images has become a preferred method for mapping land cover, especially over large areas. This study was designed to map the land cover and agricultural fields of a large area using Sentinel-1A synthetic aperture radar (SAR) images. Seven machine-learning methods were employed for image analyses. The Random Forest classifier algorithm outperformed the other machine-learning methods in the training step; thus, we selected it for further use and tuned its parameters. After several image processing steps, we classified the final image into 23 land cover classes and achieved an overall accuracy of 42% for all classes, and 57% for agricultural fields. This research note highlights some characteristics and advantages of using Sentinel-1A images and provides novel methods for nation-wide large-area mapping applications.

AB - Land use and land cover maps are vital sources of information for many applications. Recently, using high-resolution and open-access satellite images has become a preferred method for mapping land cover, especially over large areas. This study was designed to map the land cover and agricultural fields of a large area using Sentinel-1A synthetic aperture radar (SAR) images. Seven machine-learning methods were employed for image analyses. The Random Forest classifier algorithm outperformed the other machine-learning methods in the training step; thus, we selected it for further use and tuned its parameters. After several image processing steps, we classified the final image into 23 land cover classes and achieved an overall accuracy of 42% for all classes, and 57% for agricultural fields. This research note highlights some characteristics and advantages of using Sentinel-1A images and provides novel methods for nation-wide large-area mapping applications.

KW - radar

KW - land cover

UR - http://www.scopus.com/inward/record.url?scp=85068892294&partnerID=8YFLogxK

U2 - 10.1080/07038992.2019.1635877

DO - 10.1080/07038992.2019.1635877

M3 - Article

VL - 45

SP - 163

EP - 175

JO - Canadian Journal of Remote Sensing

JF - Canadian Journal of Remote Sensing

SN - 0703-8992

IS - 2

ER -