Investigating the Utility of Oblique Tree-Based Ensembles for the Classification of Hyperspectral Data

• Published in: Sensors (Basel, Switzerland) Vol. 16; no. 11; p. 1918
• Main Authors: Poona, Nitesh, van Niekerk, Adriaan, Ismail, Riyad
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 15-11-2016; MDPI
• Subjects: Accuracy; Algorithms; Asymptomatic; Band spectra; Classification; Crop diseases; hyperspectral data; oblique tree-based ensembles; Physiology; Pinus radiata; Principal components analysis; Remote sensing; Resampling; Sensors; Spectral bands; spectral resampling; Spectral resolution; Spectroscopy; Sugarcane; Support vector machines; hyperspectral data; oblique tree-based ensembles; spectral resampling; Pinus radiata
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s16111918

Ensemble 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
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
seedlings.