Please use this identifier to cite or link to this item: https://hdl.handle.net/10321/1137
DC FieldValueLanguage
dc.contributor.authorMarwala, T.en_US
dc.contributor.authorAbe, B. T.en_US
dc.contributor.authorOlugbara, Oludayo O.en_US
dc.date.accessioned2014-08-14T06:40:51Z-
dc.date.available2014-08-14T06:40:51Z-
dc.date.issued2014-06-12-
dc.identifier.citationAbe, B.T.; Olugbara, O.O. and Marwala, T. 2014. Experimental comparison of support vector machines with random forests for hyperspectral image land cover classification. Journal of Earth System Science. 123 (4): 779-790en_US
dc.identifier.issn0253-4126-
dc.identifier.issn0973-774X-
dc.identifier.urihttp://hdl.handle.net/10321/1137-
dc.description.abstractThe performances of regular support vector machines and random forests are experimentally com-pared for hyperspectral imaging land cover classification. Special characteristics of hyperspectral imaging dataset present diverse processing problems to be resolved under robust mathematical formalisms such as image classification. As a result, pixel purity index algorithm is used to obtain endmember spectral responses from Indiana pine hyperspectral image dataset. The generalized reduced gradient optimiza-tion algorithm is thereafter executed on the research data to estimate fractional abundances in the hyperspectral image and thereby obtain the numeric values for land cover classification. The Waikato environment for knowledge analysis (WEKA) data mining framework is selected as a tool to carry out the classification process by using support vector machines and random forests classifiers. Results show that performance of support vector machines is comparable to that of random forests. This study makes a positive contribution to the problem of land cover classification by exploring generalized reduced gra-dient method, support vector machines, and random forests to improve producer accuracy and overall classification accuracy. The performance comparison of these classifiers is valuable for a decision maker to consider tradeoffs in method accuracy versus method complexity.en_US
dc.format.extent12 pen_US
dc.language.isoenen_US
dc.publisherIndian Academy of Sciencesen_US
dc.relation.ispartofJournal of earth system science (Online)en_US
dc.subjectHyperspectral imageen_US
dc.subjectLand coveren_US
dc.subjectGeneralized reduced gradienten_US
dc.subjectClassifiersen_US
dc.titleExperimental comparison of support vector machines with random forests for hyperspectral image land cover classificationen_US
dc.typeArticleen_US
dc.publisher.urihttp://download.springer.com/static/pdf/252/art%253A10.1007%252Fs12040-014-0436-x.pdf?auth66=1408093933_acb520a22c71dc468fdc1783b80c6c0b&ext=.pdfen_US
dc.dut-rims.pubnumDUT-004283en_US
dc.identifier.doi10.1007/s12040-014-0436-x-
local.sdgSDG15-
local.sdgSDG14-
item.grantfulltextopen-
item.cerifentitytypePublications-
item.fulltextWith Fulltext-
item.openairecristypehttp://purl.org/coar/resource_type/c_18cf-
item.languageiso639-1en-
item.openairetypeArticle-
Appears in Collections:Research Publications (Accounting and Informatics)
Files in This Item:
File Description SizeFormat
olugbara_et_al_j_earth_syst_sci_2014.pdf1.55 MBAdobe PDFThumbnail
View/Open
Show simple item record

Page view(s) 50

772
checked on Dec 22, 2024

Download(s) 50

458
checked on Dec 22, 2024

Google ScholarTM

Check

Altmetric

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.