Change detection from satellite data time series using pixel value distributions

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

Abstract

The paper proposes a novel approach for change detection from image time series. In this approach changes are detected from evaluated distances between the (possibly multivariate) distributions of pixel values. Basing change detection on these distributions facilitates, e.g., joint analysis of images having different resolutions and comparisons of smaller areas against larger images. Furthermore, clouded areas can be excluded from each image separately, allowing the data in the remaining pixels to be utilized independent on the whether the corresponding pixels have been covered by clouds in the other images in the time series. In the paper the proposed method is applied to forest cover change detection using Landsat data covering Mexico.
Original languageEnglish
Title of host publicationProceedings of ESA Living Planet Symposium
PublisherEuropean Space Agency ESA
ISBN (Print)978-9-2922-1286-5
Publication statusPublished - 2013
MoE publication typeNot Eligible
EventESA Living Planet Symposium 2013 - Edinburgh, United Kingdom
Duration: 9 Sep 201313 Sep 2013
Conference number: ESA SP-722

Conference

ConferenceESA Living Planet Symposium 2013
CountryUnited Kingdom
CityEdinburgh
Period9/09/1313/09/13

Fingerprint

satellite data
pixel
time series
forest cover
Landsat
detection
distribution

Keywords

  • time series
  • change detection

Cite this

Saarela, O., Molinier, M., & Sirro, L. (2013). Change detection from satellite data time series using pixel value distributions. In Proceedings of ESA Living Planet Symposium European Space Agency ESA.
@inproceedings{5ec1f4960e1b46159e486e3ff9920135,
title = "Change detection from satellite data time series using pixel value distributions",
abstract = "The paper proposes a novel approach for change detection from image time series. In this approach changes are detected from evaluated distances between the (possibly multivariate) distributions of pixel values. Basing change detection on these distributions facilitates, e.g., joint analysis of images having different resolutions and comparisons of smaller areas against larger images. Furthermore, clouded areas can be excluded from each image separately, allowing the data in the remaining pixels to be utilized independent on the whether the corresponding pixels have been covered by clouds in the other images in the time series. In the paper the proposed method is applied to forest cover change detection using Landsat data covering Mexico.",
keywords = "time series, change detection",
author = "Olli Saarela and Matthieu Molinier and Laura Sirro",
note = "Project code: 40736",
year = "2013",
language = "English",
isbn = "978-9-2922-1286-5",
booktitle = "Proceedings of ESA Living Planet Symposium",
publisher = "European Space Agency ESA",
address = "France",

}

Saarela, O, Molinier, M & Sirro, L 2013, Change detection from satellite data time series using pixel value distributions. in Proceedings of ESA Living Planet Symposium. European Space Agency ESA, ESA Living Planet Symposium 2013, Edinburgh, United Kingdom, 9/09/13.

Change detection from satellite data time series using pixel value distributions. / Saarela, Olli; Molinier, Matthieu; Sirro, Laura.

Proceedings of ESA Living Planet Symposium. European Space Agency ESA, 2013.

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

TY - GEN

T1 - Change detection from satellite data time series using pixel value distributions

AU - Saarela, Olli

AU - Molinier, Matthieu

AU - Sirro, Laura

N1 - Project code: 40736

PY - 2013

Y1 - 2013

N2 - The paper proposes a novel approach for change detection from image time series. In this approach changes are detected from evaluated distances between the (possibly multivariate) distributions of pixel values. Basing change detection on these distributions facilitates, e.g., joint analysis of images having different resolutions and comparisons of smaller areas against larger images. Furthermore, clouded areas can be excluded from each image separately, allowing the data in the remaining pixels to be utilized independent on the whether the corresponding pixels have been covered by clouds in the other images in the time series. In the paper the proposed method is applied to forest cover change detection using Landsat data covering Mexico.

AB - The paper proposes a novel approach for change detection from image time series. In this approach changes are detected from evaluated distances between the (possibly multivariate) distributions of pixel values. Basing change detection on these distributions facilitates, e.g., joint analysis of images having different resolutions and comparisons of smaller areas against larger images. Furthermore, clouded areas can be excluded from each image separately, allowing the data in the remaining pixels to be utilized independent on the whether the corresponding pixels have been covered by clouds in the other images in the time series. In the paper the proposed method is applied to forest cover change detection using Landsat data covering Mexico.

KW - time series

KW - change detection

M3 - Conference article in proceedings

SN - 978-9-2922-1286-5

BT - Proceedings of ESA Living Planet Symposium

PB - European Space Agency ESA

ER -

Saarela O, Molinier M, Sirro L. Change detection from satellite data time series using pixel value distributions. In Proceedings of ESA Living Planet Symposium. European Space Agency ESA. 2013