The University of Southampton
University of Southampton Institutional Repository

Incorporating spatial variability measures in land-cover classification using random forest

Incorporating spatial variability measures in land-cover classification using random forest
Incorporating spatial variability measures in land-cover classification using random forest
The spatialvariability of remotely sensed image values provides important information about the arrangement of objects and their spatial relationships within the image. The characterisation of spatialvariability in such images, for example, to measure of texture, is of great utility for the discrimination of landcover classes. To this end, the variogram, a function commonly applied in geostatistics, has been used widely to extract image texture for remotely sensed data classification.

The aim of this study was to assess the increase in accuracy that can be achieved by incorporating univariate and multivariate textural measures of Landsat TM imagery in classification models applied to large heterogeneous landscapes. Such landscapes which difficult to classify due to the large number of landcover categories and low inter-class separability. Madogram, rodogram and direct variogram for the univariate case, and cross- and pseudocross variograms for the multivariate one, together with multi-seasonal spectral information were used in a RandomForest classifier to map landcover types.

The addition of spatialvariability into multi-seasonal RandomForest models leads to an increase in the overall accuracy of 8%, and to an increase in the Kappa index of 9%, respectively. The increase in per categories Kappa for the textural RandomForest model reached 30% for certain categories. This study demonstrates that the use of information on spatialvariability produces a fundamental increase in per class classification accuracy of complex land-cover categories
1878-0296
44-49
Rodriguez-Galiano, V.F.
1eb6a1dd-f73d-4e90-a9cf-a51f20712c3c
Abarca-Hernandez, F.
287e84cf-6fd0-4a81-ad6e-729318b7749b
Ghimire, B.
d2ab9f23-815a-485e-8347-15f28ef8ce77
Chica-Olmo, M.
c7291c15-3b53-45d7-942c-06985f77d6f6
Atkinson, P.M.
96e96579-56fe-424d-a21c-17b6eed13b0b
Jeganathan, C.
2dd3d151-7bb1-42b9-bda6-54df0bed4ae8
Rodriguez-Galiano, V.F.
1eb6a1dd-f73d-4e90-a9cf-a51f20712c3c
Abarca-Hernandez, F.
287e84cf-6fd0-4a81-ad6e-729318b7749b
Ghimire, B.
d2ab9f23-815a-485e-8347-15f28ef8ce77
Chica-Olmo, M.
c7291c15-3b53-45d7-942c-06985f77d6f6
Atkinson, P.M.
96e96579-56fe-424d-a21c-17b6eed13b0b
Jeganathan, C.
2dd3d151-7bb1-42b9-bda6-54df0bed4ae8

Rodriguez-Galiano, V.F., Abarca-Hernandez, F., Ghimire, B., Chica-Olmo, M., Atkinson, P.M. and Jeganathan, C. (2011) Incorporating spatial variability measures in land-cover classification using random forest. Procedia Environmental Sciences, 3, 44-49. (doi:10.1016/j.proenv.2011.02.009).

Record type: Article

Abstract

The spatialvariability of remotely sensed image values provides important information about the arrangement of objects and their spatial relationships within the image. The characterisation of spatialvariability in such images, for example, to measure of texture, is of great utility for the discrimination of landcover classes. To this end, the variogram, a function commonly applied in geostatistics, has been used widely to extract image texture for remotely sensed data classification.

The aim of this study was to assess the increase in accuracy that can be achieved by incorporating univariate and multivariate textural measures of Landsat TM imagery in classification models applied to large heterogeneous landscapes. Such landscapes which difficult to classify due to the large number of landcover categories and low inter-class separability. Madogram, rodogram and direct variogram for the univariate case, and cross- and pseudocross variograms for the multivariate one, together with multi-seasonal spectral information were used in a RandomForest classifier to map landcover types.

The addition of spatialvariability into multi-seasonal RandomForest models leads to an increase in the overall accuracy of 8%, and to an increase in the Kappa index of 9%, respectively. The increase in per categories Kappa for the textural RandomForest model reached 30% for certain categories. This study demonstrates that the use of information on spatialvariability produces a fundamental increase in per class classification accuracy of complex land-cover categories

This record has no associated files available for download.

More information

Published date: 2011
Additional Information: 1st Conference on Spatial Statistics 2011 – Mapping Global Change
Organisations: Global Env Change & Earth Observation

Identifiers

Local EPrints ID: 343288
URI: http://eprints.soton.ac.uk/id/eprint/343288
ISSN: 1878-0296
PURE UUID: 54b7bf41-1a24-4e74-bbf2-b098ed14a0a7
ORCID for P.M. Atkinson: ORCID iD orcid.org/0000-0002-5489-6880

Catalogue record

Date deposited: 03 Oct 2012 08:55
Last modified: 15 Mar 2024 02:47

Export record

Altmetrics

Contributors

Author: V.F. Rodriguez-Galiano
Author: F. Abarca-Hernandez
Author: B. Ghimire
Author: M. Chica-Olmo
Author: P.M. Atkinson ORCID iD
Author: C. Jeganathan

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×