ITEM VIEW

Transferability of decision trees for land cover classification in a heterogeneous area

dc.contributor.authorVerhulp, Julie Katherineen_ZA
dc.contributor.authorVan Niekerk, Adriaanen_ZA
dc.date.accessioned2018-12-10T09:59:19Z
dc.date.available2018-12-10T09:59:19Z
dc.date.issued2017-04
dc.identifier.citationVerhulp, J. & Van Niekerk, A. 2017. Transferability of decision trees for land cover classification in a heterogeneous area. South African Journal of Geomatics, 6(1):30-46, doi:10.4314/sajg.v6i1.3en_ZA
dc.identifier.issn2225-8531 (online)
dc.identifier.otherdoi:10.4314/sajg.v6i1.3
dc.identifier.urihttp://hdl.handle.net/10019.1/105246en_ZA
dc.descriptionCITATION: Verhulp, J. & Van Niekerk, A. 2017. Transferability of decision trees for land cover classification in a heterogeneous area. South African Journal of Geomatics, 6(1):30-46, doi:10.4314/sajg.v6i1.3.en_ZA
dc.descriptionThe original publication is available at http://www.sajg.org.zaen_ZA
dc.description.abstractAs the value of accurate land cover becomes more apparent, methods to decrease the costs associated with supervised land cover mapping are investigated. One such method is to use training data captured in one scene and apply it to a different scene through a process known as signature extension. This paper attempts to derive classification rules from training data of four Landsat-8 scenes by using the classification and regression tree (CART) implementation of the decision tree algorithm. The transferability of the ruleset was evaluated by classifying two adjacent scenes. The classification of the four mosaicked scenes achieved an overall accuracy of 80.6%, while the two adjacent scenes achieved 61.4% and 83.7% respectively. The low accuracy of the first adjacent scene can be ascribed to a misclassification of graminoids, urban and bare areas, attributed to the temporal changes of grasslands throughout the year. In an attempt to improve the results, a normalised difference vegetation index (NDVI) threshold was applied to each scene. This increased the accuracy of the first adjacent scene but decreased the accuracy of the second. We conclude that signature extension using CART is unreliable. However, simple rules can be added to improve the results.en_ZA
dc.description.urihttp://www.sajg.org.za/index.php/sajg/article/view/438
dc.format.extent17 pages : illustrationsen_ZA
dc.language.isoen_ZAen_ZA
dc.publisherCONSASen_ZA
dc.subjectTreesen_ZA
dc.subjectLand cover mappingen_ZA
dc.subjectSupervised classificationen_ZA
dc.subjectDecision trees -- Learningen_ZA
dc.subjectClassification and Regression Tree (CART)
dc.subjectClassifier extensionen_ZA
dc.subjectRemote sensingen_ZA
dc.subjectLandsat-8en_ZA
dc.subjectLearning Classifier Systemsen_ZA
dc.subjectLand use -- Remote sensingen_ZA
dc.subjectLand use -- Research -- Methodologyen_ZA
dc.subjectCartography -- Remote sensingen_ZA
dc.subjectNormalised Difference Vegetation Index (NDVI)en_ZA
dc.titleTransferability of decision trees for land cover classification in a heterogeneous areaen_ZA
dc.typeArticleen_ZA
dc.description.versionPublisher's versionen_ZA
dc.rights.holderAuthors retain copyrighten_ZA


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

ITEM VIEW