Investigating the utility of oblique tree-based ensembles for the classification of hyperspectral data

Simple item page
dc.contributor.authorPoona, Nitesh Keshavelalen_ZA
dc.contributor.authorVan Niekerk, Adriaanen_ZA
dc.contributor.authorIsmail, Riyaden_ZA
dc.date.accessioned2016-11-30T14:04:51Z
dc.date.available2016-11-30T14:04:51Z
dc.date.issued2016-11
dc.descriptionCITATION: Poona, N., Van Niekerk, A. & Ismail, R. 2016. Investigating the utility of oblique tree-based ensembles for the classification of hyperspectral data. Sensors, 16(11):1918, doi:10.3390/s16111918.en_ZA
dc.descriptionThe original publication is available at www.mdpi.comen_ZA
dc.descriptionPublication of this article was funded by the Stellenbosch University Open Access Fund.en_ZA
dc.description.abstractEnsemble classifiers are being widely used for the classification of spectroscopic data. In this regard, the random forest (RF) ensemble has been successfully applied in an array of applications, and has proven to be robust in handling high dimensional data. More recently, several variants of the traditional RF algorithm including rotation forest (rotF) and oblique random forest (oRF) have been applied to classifying high dimensional data. In this study we compare the traditional RF, rotF, and oRF (using three different splitting rules, i.e., ridge regression, partial least squares, and support vector machine) for the classification of healthy and infected Pinus radiata seedlings using high dimensional spectroscopic data. We further test the robustness of these five ensemble classifiers to reduced spectral resolution by spectral resampling (binning) of the original spectral bands. The results showed that the three oblique random forest ensembles outperformed both the traditional RF and rotF ensembles. Additionally, the rotF ensemble proved to be the least robust of the five ensembles tested. Spectral resampling of the original bands provided mixed results. Nevertheless, the results demonstrate that using spectral resampled bands is a promising approach to classifying asymptomatic stress in Pinus radiata seedlings.en_ZA
dc.description.urihttp://www.mdpi.com/1424-8220/16/11/1918
dc.description.versionPublisher's version
dc.format.extent16 pages
dc.identifier.citationPoona, N., Van Niekerk, A. & Ismail, R. 2016. Investigating the utility of oblique tree-based ensembles for the classification of hyperspectral data. Sensors, 16(11):1918, doi:10.3390/s16111918en_ZA
dc.identifier.issn1424-8220 (online)
dc.identifier.otherdoi:10.3390/s16111918
dc.identifier.urihttp://hdl.handle.net/10019.1/99925
dc.language.isoen_ZAen_ZA
dc.publisherMDPI
dc.rights.holderAuthors retain copyright
dc.subjectSpectroscopy -- Imagingen_ZA
dc.subjectOblique tree-based ensemblesen_ZA
dc.subjectPinus radiataen_ZA
dc.subjectSensorsen_ZA
dc.titleInvestigating the utility of oblique tree-based ensembles for the classification of hyperspectral dataen_ZA
dc.typeArticleen_ZA
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
poona_investigating_2016.pdf
Size:
1.1 MB
Format:
Adobe Portable Document Format
Description:
Download article
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.95 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Research Articles (Geography and Environmental Studies)