Adjorlolo, CMutanga, OCho, Moses AIsmail, R2013-01-282013-01-282013-04Adjorlolo, C., Mutanga, O., Cho, M.A. and Ismail, R. 2013. Spectral resampling based on user-defined interband correlation filter: C3 and C4 grass species classification. International Journal of Applied Earth Observation and Geoinformation Science, vol. 21, pp. 535-5440303-2434http://www.sciencedirect.com/science/article/pii/S0303243412001493http://hdl.handle.net/10204/6447Copyright: 2012 Elsevier. This is the preprint version of the work. The definitive version is published in International Journal of Applied Earth Observation and Geoinformation Science, vol. 21, pp. 535-544In this paper, a user-defined inter-band correlation filter function was used to resample hyperspectral data and thereby mitigate the problem of multicollinearity in classification analysis. The proposed resampling technique convolves the spectral dependence information between a chosen band-centre and its shorter and longer wavelength neighbours. Weighting threshold of inter-band correlation (WTC, Pearson’s r) was calculated, whereby r = 1 at the band-centre. Various WTC (r = 0.99, r = 0.95 and r = 0.90) were assessed, and bands with 0 coefficients beyond a chosen threshold were assigned r = 0. The resultant data were used in the random forest analysis to classify C3 and C4 grass species. The respective WTC datasets yielded improved classification accuracies (kappa = 0.82, 0.79 and 0.76) with less correlated wavebands when compared to resampled Hyperion bands (kappa = 0.76). Overall, the results obtained from this study suggested that resampling of hyperspectral data should account for the spectral dependence information to improve overall classification accuracy as well as reducing the problem of multicollinearity.enEarth observationHyperspectral dataSpectral resamplingInter-band correlationGrass species classificationRandom forestsArticleAdjorlolo, C., Mutanga, O., Cho, M. A., & Ismail, R. (2013). Spectral resampling based on user-defined interband correlation filter: C3 and C4 grass species classification. http://hdl.handle.net/10204/6447Adjorlolo, C, O Mutanga, Moses A Cho, and R Ismail "Spectral resampling based on user-defined interband correlation filter: C3 and C4 grass species classification." (2013) http://hdl.handle.net/10204/6447Adjorlolo C, Mutanga O, Cho MA, Ismail R. Spectral resampling based on user-defined interband correlation filter: C3 and C4 grass species classification. 2013; http://hdl.handle.net/10204/6447.TY - Article AU - Adjorlolo, C AU - Mutanga, O AU - Cho, Moses A AU - Ismail, R AB - In this paper, a user-defined inter-band correlation filter function was used to resample hyperspectral data and thereby mitigate the problem of multicollinearity in classification analysis. The proposed resampling technique convolves the spectral dependence information between a chosen band-centre and its shorter and longer wavelength neighbours. Weighting threshold of inter-band correlation (WTC, Pearson’s r) was calculated, whereby r = 1 at the band-centre. Various WTC (r = 0.99, r = 0.95 and r = 0.90) were assessed, and bands with 0 coefficients beyond a chosen threshold were assigned r = 0. The resultant data were used in the random forest analysis to classify C3 and C4 grass species. The respective WTC datasets yielded improved classification accuracies (kappa = 0.82, 0.79 and 0.76) with less correlated wavebands when compared to resampled Hyperion bands (kappa = 0.76). Overall, the results obtained from this study suggested that resampling of hyperspectral data should account for the spectral dependence information to improve overall classification accuracy as well as reducing the problem of multicollinearity. DA - 2013-04 DB - ResearchSpace DP - CSIR KW - Earth observation KW - Hyperspectral data KW - Spectral resampling KW - Inter-band correlation KW - Grass species classification KW - Random forests LK - https://researchspace.csir.co.za PY - 2013 SM - 0303-2434 T1 - Spectral resampling based on user-defined interband correlation filter: C3 and C4 grass species classification TI - Spectral resampling based on user-defined interband correlation filter: C3 and C4 grass species classification UR - http://hdl.handle.net/10204/6447 ER -