Polarimetric SAR data in land cover mapping in boreal zone

Research output: Contribution to journalArticleScientificpeer-review

40 Citations (Scopus)

Abstract

This paper compares ALOS PALSAR fully polarimetric and dual-polarized data in the application area of land cover mapping. To assure versatile comparison of the data, different classification methods and different features of data are used. Two of the classification methods used are based on supervised classification and two on unsupervised classification. Polarimetric data are used in three ways: 1) as fully polarimetric data; 2) features calculated from fully polarimetric data; and 3) intensity data of selected channels. Combinations of six (water, field, sparse forest, dense forest, peat land, and urban areas), five, four, and three classes were used for classification. Fully polarimetric data gave better results (87.5%84.7% with three classes; open land areas, forest, and water) than intensity data only (83.6%78.6%), but the differences in the overall accuracies between the methods were not more than 7.6%. Kappa coefficients of agreement are moderate for all the classifications. Supervised classification can be expected to perform better than unsupervised classification, given that the training areas can be selected accurately. Dual polarization data were found to be an attractive alternative in cases where fully polarimetric data are not available or it is of low resolution. With intensities of selected polarimetric features, it was possible to obtain a high classification accuracy as with fully polarimetric data. This also opens possibilities for nonspecialist users to benefit from polarimetric information in classification.

Original languageEnglish
Article number5475228
Pages (from-to)3652-3662
Number of pages11
JournalIEEE Transactions on Geoscience and Remote Sensing
Volume48
Issue number10
DOIs
Publication statusPublished - 1 Oct 2010
MoE publication typeA1 Journal article-refereed

Fingerprint

land cover
synthetic aperture radar
unsupervised classification
image classification
Peat
PALSAR
ALOS
Water
peat
Polarization
polarization
urban area
water

Keywords

  • Advanced Land Observing Satellite Phased Array type L-band Synthetic Aperture Radar (ALOS PALSAR)
  • classification
  • land cover
  • polarimetric synthetic aperture radar

Cite this

@article{791d5a5d696b4dd89fbd5a03498a3e7c,
title = "Polarimetric SAR data in land cover mapping in boreal zone",
abstract = "This paper compares ALOS PALSAR fully polarimetric and dual-polarized data in the application area of land cover mapping. To assure versatile comparison of the data, different classification methods and different features of data are used. Two of the classification methods used are based on supervised classification and two on unsupervised classification. Polarimetric data are used in three ways: 1) as fully polarimetric data; 2) features calculated from fully polarimetric data; and 3) intensity data of selected channels. Combinations of six (water, field, sparse forest, dense forest, peat land, and urban areas), five, four, and three classes were used for classification. Fully polarimetric data gave better results (87.5{\%}84.7{\%} with three classes; open land areas, forest, and water) than intensity data only (83.6{\%}78.6{\%}), but the differences in the overall accuracies between the methods were not more than 7.6{\%}. Kappa coefficients of agreement are moderate for all the classifications. Supervised classification can be expected to perform better than unsupervised classification, given that the training areas can be selected accurately. Dual polarization data were found to be an attractive alternative in cases where fully polarimetric data are not available or it is of low resolution. With intensities of selected polarimetric features, it was possible to obtain a high classification accuracy as with fully polarimetric data. This also opens possibilities for nonspecialist users to benefit from polarimetric information in classification.",
keywords = "Advanced Land Observing Satellite Phased Array type L-band Synthetic Aperture Radar (ALOS PALSAR), classification, land cover, polarimetric synthetic aperture radar",
author = "Anne L{\"o}nnqvist and Yrj{\"o} Rauste and Matthieu Molinier and Tuomas H{\"a}me",
year = "2010",
month = "10",
day = "1",
doi = "10.1109/TGRS.2010.2048115",
language = "English",
volume = "48",
pages = "3652--3662",
journal = "IEEE Transactions on Geoscience and Remote Sensing",
issn = "0196-2892",
publisher = "Institute of Electrical and Electronic Engineers IEEE",
number = "10",

}

Polarimetric SAR data in land cover mapping in boreal zone. / Lönnqvist, Anne; Rauste, Yrjö; Molinier, Matthieu; Häme, Tuomas.

In: IEEE Transactions on Geoscience and Remote Sensing, Vol. 48, No. 10, 5475228, 01.10.2010, p. 3652-3662.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Polarimetric SAR data in land cover mapping in boreal zone

AU - Lönnqvist, Anne

AU - Rauste, Yrjö

AU - Molinier, Matthieu

AU - Häme, Tuomas

PY - 2010/10/1

Y1 - 2010/10/1

N2 - This paper compares ALOS PALSAR fully polarimetric and dual-polarized data in the application area of land cover mapping. To assure versatile comparison of the data, different classification methods and different features of data are used. Two of the classification methods used are based on supervised classification and two on unsupervised classification. Polarimetric data are used in three ways: 1) as fully polarimetric data; 2) features calculated from fully polarimetric data; and 3) intensity data of selected channels. Combinations of six (water, field, sparse forest, dense forest, peat land, and urban areas), five, four, and three classes were used for classification. Fully polarimetric data gave better results (87.5%84.7% with three classes; open land areas, forest, and water) than intensity data only (83.6%78.6%), but the differences in the overall accuracies between the methods were not more than 7.6%. Kappa coefficients of agreement are moderate for all the classifications. Supervised classification can be expected to perform better than unsupervised classification, given that the training areas can be selected accurately. Dual polarization data were found to be an attractive alternative in cases where fully polarimetric data are not available or it is of low resolution. With intensities of selected polarimetric features, it was possible to obtain a high classification accuracy as with fully polarimetric data. This also opens possibilities for nonspecialist users to benefit from polarimetric information in classification.

AB - This paper compares ALOS PALSAR fully polarimetric and dual-polarized data in the application area of land cover mapping. To assure versatile comparison of the data, different classification methods and different features of data are used. Two of the classification methods used are based on supervised classification and two on unsupervised classification. Polarimetric data are used in three ways: 1) as fully polarimetric data; 2) features calculated from fully polarimetric data; and 3) intensity data of selected channels. Combinations of six (water, field, sparse forest, dense forest, peat land, and urban areas), five, four, and three classes were used for classification. Fully polarimetric data gave better results (87.5%84.7% with three classes; open land areas, forest, and water) than intensity data only (83.6%78.6%), but the differences in the overall accuracies between the methods were not more than 7.6%. Kappa coefficients of agreement are moderate for all the classifications. Supervised classification can be expected to perform better than unsupervised classification, given that the training areas can be selected accurately. Dual polarization data were found to be an attractive alternative in cases where fully polarimetric data are not available or it is of low resolution. With intensities of selected polarimetric features, it was possible to obtain a high classification accuracy as with fully polarimetric data. This also opens possibilities for nonspecialist users to benefit from polarimetric information in classification.

KW - Advanced Land Observing Satellite Phased Array type L-band Synthetic Aperture Radar (ALOS PALSAR)

KW - classification

KW - land cover

KW - polarimetric synthetic aperture radar

UR - http://www.scopus.com/inward/record.url?scp=77957005457&partnerID=8YFLogxK

U2 - 10.1109/TGRS.2010.2048115

DO - 10.1109/TGRS.2010.2048115

M3 - Article

AN - SCOPUS:77957005457

VL - 48

SP - 3652

EP - 3662

JO - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

SN - 0196-2892

IS - 10

M1 - 5475228

ER -