File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: The first all-season sample set for mapping global land cover with Landsat-8 data

TitleThe first all-season sample set for mapping global land cover with Landsat-8 data
Authors
KeywordsAnytime
Validation
Latitudinal zones
Training sample
Anywhere
Issue Date2017
Citation
Science Bulletin, 2017, v. 62, n. 7, p. 508-515 How to Cite?
Abstract© 2017 Science China Press We report the world's first all-season training and validation sample sets for global land cover classification with Landsat-8 data. Prior to this, such samples were only available at a single date primarily from the growing season. It is unknown how much limitation such a single-date sample has to mapping global land cover in other seasons of the year. To answer this question, we selected available Landsat-8 images from four seasons and collected training and validation samples from them. We compared the performances of training samples in different seasons using Random Forest algorithm. We found that the use of training samples from any individual season would result in the best overall classification accuracy when validated by samples in the same season. The global overall accuracy from combined best seasonal results was 67.2% when classifying the 11 Level-1 classes in the Finer Resolution Observation and Monitoring of Global Land Cover (FROM-GLC) classification system. The use of training samples from all seasons (named all-season training sample set hereafter) produced an overall accuracy of 67.0%. We also tested classification within 10° latitude 60° longitude zones using all-season training subsample within each zone and obtained an overall accuracy of 70.2%. This indicates that properly grouped subsamples in space can help improve classification accuracies. All the results in this study seem to suggest that it is possible to use an all-season training sample set to reach global optimality with universal applicability in classifying images acquired at any time of a year for global land cover mapping.
Persistent Identifierhttp://hdl.handle.net/10722/296818
ISSN
2020 Impact Factor: 11.78
2020 SCImago Journal Rankings: 1.983
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorLi, Congcong-
dc.contributor.authorGong, Peng-
dc.contributor.authorWang, Jie-
dc.contributor.authorZhu, Zhiliang-
dc.contributor.authorBiging, Gregory S.-
dc.contributor.authorYuan, Cui-
dc.contributor.authorHu, Tengyun-
dc.contributor.authorZhang, Haiying-
dc.contributor.authorWang, Qi-
dc.contributor.authorLi, Xuecao-
dc.contributor.authorLiu, Xiaoxuan-
dc.contributor.authorXu, Yidi-
dc.contributor.authorGuo, Jing-
dc.contributor.authorLiu, Caixia-
dc.contributor.authorHackman, Kwame O.-
dc.contributor.authorZhang, Meinan-
dc.contributor.authorCheng, Yuqi-
dc.contributor.authorYu, Le-
dc.contributor.authorYang, Jun-
dc.contributor.authorHuang, Huabing-
dc.contributor.authorClinton, Nicholas-
dc.date.accessioned2021-02-25T15:16:45Z-
dc.date.available2021-02-25T15:16:45Z-
dc.date.issued2017-
dc.identifier.citationScience Bulletin, 2017, v. 62, n. 7, p. 508-515-
dc.identifier.issn2095-9273-
dc.identifier.urihttp://hdl.handle.net/10722/296818-
dc.description.abstract© 2017 Science China Press We report the world's first all-season training and validation sample sets for global land cover classification with Landsat-8 data. Prior to this, such samples were only available at a single date primarily from the growing season. It is unknown how much limitation such a single-date sample has to mapping global land cover in other seasons of the year. To answer this question, we selected available Landsat-8 images from four seasons and collected training and validation samples from them. We compared the performances of training samples in different seasons using Random Forest algorithm. We found that the use of training samples from any individual season would result in the best overall classification accuracy when validated by samples in the same season. The global overall accuracy from combined best seasonal results was 67.2% when classifying the 11 Level-1 classes in the Finer Resolution Observation and Monitoring of Global Land Cover (FROM-GLC) classification system. The use of training samples from all seasons (named all-season training sample set hereafter) produced an overall accuracy of 67.0%. We also tested classification within 10° latitude 60° longitude zones using all-season training subsample within each zone and obtained an overall accuracy of 70.2%. This indicates that properly grouped subsamples in space can help improve classification accuracies. All the results in this study seem to suggest that it is possible to use an all-season training sample set to reach global optimality with universal applicability in classifying images acquired at any time of a year for global land cover mapping.-
dc.languageeng-
dc.relation.ispartofScience Bulletin-
dc.subjectAnytime-
dc.subjectValidation-
dc.subjectLatitudinal zones-
dc.subjectTraining sample-
dc.subjectAnywhere-
dc.titleThe first all-season sample set for mapping global land cover with Landsat-8 data-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1016/j.scib.2017.03.011-
dc.identifier.scopuseid_2-s2.0-85019184935-
dc.identifier.volume62-
dc.identifier.issue7-
dc.identifier.spage508-
dc.identifier.epage515-
dc.identifier.eissn2095-9281-
dc.identifier.isiWOS:000403560800012-
dc.identifier.issnl2095-9273-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats