File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Exploring the potential role of feature selection in global land-cover mapping

TitleExploring the potential role of feature selection in global land-cover mapping
Authors
Issue Date2016
Citation
International Journal of Remote Sensing, 2016, v. 37, n. 23, p. 5491-5504 How to Cite?
Abstract© 2016 Informa UK Limited, trading as Taylor & Francis Group. Global land cover has been acknowledged as a fundamental variable in several global-scale studies for environment and climate change. Recent developments in global land-cover mapping focused on spatial resolution improvement with more heterogeneous features to integrate the spatial, spectral, and temporal information. Although the high dimensional input features as a whole lead to discriminatory strengths to produce more accurate land-cover maps, it comes at the cost of an increased classification complexity. The feature selection method has become a necessity for dimensionality reduction in classification with large amounts of input features. In this study, the potential of feature selection in global land-cover mapping is explored. A total of 63 features derived from the Landsat Thematic Mapper (TM) spectral bands, Moderate Resolution Imaging Spectroradiometer (MODIS) time series enhanced vegetation index (EVI) data, digital elevation model (DEM), and many climate-ecological variables and global training samples are input to k-nearest neighbours (k-NN) and Random Forest (RF) classifiers. Two filter feature selection algorithms, i.e. Relieff and max-min-associated (MNA), were employed to select the optimal subsets of features for the whole world and different biomes. The mapping accuracies with/without feature selection were evaluated by a global validation sample set. Overall, the result indicates no significant accuracy improvement in global land-cover mapping after dimensionality reduction. Nevertheless, feature selection has the capability of identifying useful features in different biomes and improves the computational efficiency, which is valuable in global-scale computing.
Persistent Identifierhttp://hdl.handle.net/10722/296801
ISSN
2023 Impact Factor: 3.0
2023 SCImago Journal Rankings: 0.776
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorYu, Le-
dc.contributor.authorFu, Haohuan-
dc.contributor.authorWu, Bo-
dc.contributor.authorClinton, Nicolas-
dc.contributor.authorGong, Peng-
dc.date.accessioned2021-02-25T15:16:42Z-
dc.date.available2021-02-25T15:16:42Z-
dc.date.issued2016-
dc.identifier.citationInternational Journal of Remote Sensing, 2016, v. 37, n. 23, p. 5491-5504-
dc.identifier.issn0143-1161-
dc.identifier.urihttp://hdl.handle.net/10722/296801-
dc.description.abstract© 2016 Informa UK Limited, trading as Taylor & Francis Group. Global land cover has been acknowledged as a fundamental variable in several global-scale studies for environment and climate change. Recent developments in global land-cover mapping focused on spatial resolution improvement with more heterogeneous features to integrate the spatial, spectral, and temporal information. Although the high dimensional input features as a whole lead to discriminatory strengths to produce more accurate land-cover maps, it comes at the cost of an increased classification complexity. The feature selection method has become a necessity for dimensionality reduction in classification with large amounts of input features. In this study, the potential of feature selection in global land-cover mapping is explored. A total of 63 features derived from the Landsat Thematic Mapper (TM) spectral bands, Moderate Resolution Imaging Spectroradiometer (MODIS) time series enhanced vegetation index (EVI) data, digital elevation model (DEM), and many climate-ecological variables and global training samples are input to k-nearest neighbours (k-NN) and Random Forest (RF) classifiers. Two filter feature selection algorithms, i.e. Relieff and max-min-associated (MNA), were employed to select the optimal subsets of features for the whole world and different biomes. The mapping accuracies with/without feature selection were evaluated by a global validation sample set. Overall, the result indicates no significant accuracy improvement in global land-cover mapping after dimensionality reduction. Nevertheless, feature selection has the capability of identifying useful features in different biomes and improves the computational efficiency, which is valuable in global-scale computing.-
dc.languageeng-
dc.relation.ispartofInternational Journal of Remote Sensing-
dc.titleExploring the potential role of feature selection in global land-cover mapping-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1080/01431161.2016.1244365-
dc.identifier.scopuseid_2-s2.0-84994000416-
dc.identifier.volume37-
dc.identifier.issue23-
dc.identifier.spage5491-
dc.identifier.epage5504-
dc.identifier.eissn1366-5901-
dc.identifier.isiWOS:000387560400001-
dc.identifier.issnl0143-1161-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats