Ensemble Learning Based Multiple Kernel Principal Component Analysis for Dimensionality Reduction and Classification of Hyperspectral Imagery

Figure 5

Classification results obtained with the KSC: (a) labeled scene and classification maps using the (b) raw data; (c) PCA; (d) KPCA_RBF; (e) KPCA_Lap; (f) KPCA_Cau; (g) KPCA_HIST; and (h) A-KPCA; (i) M-KPCA. The classification was done by SVM with an RBF kernel.

(a) Labeled scene

(b) Raw

(c) PCA

(d) KPCA_RBF

(e) KPCA_Lap

(f) KPCA_Cau

(g) KPCA_HIST

(h) A-KPCA

(i) M-KPCA