SICEM: A Generation Approach of Band Combination for Hyperspectral Imagery Reconstitution Based on Space and Information Analyses

Table 3

Classification accuracies (%) achieved using 21 selected bands on PaviaU with SVM classifier.

Class	Algorithm
	MVPCA	WaLuDi	DBSCAN	I-FDPC	SICEM
Asphalt	86.5	87.6	87.7	89.8	91.7
Meadows	96.4	96.8	96.2	94.8	96.2
Gravel	72.7	71.3	75.8	74.2	77.1
Trees	84.2	91.3	91	92.1	93.5
Painted metal sheets	99.1	99.4	99.4	97.8	99.3
Bare soil	53.2	82.5	81.5	82.4	83.8
Bitumen	79.8	83.8	81.7	76.6	83.4
Self-Blocking Bricks	83.4	85.2	86.3	83.8	85.1
Shadows	97.2	98.5	98.5	98	98.7
AA	83.6	88.4	88.6	87.7	89.9
OA	85.4	89.3	89.1	89.3	90.5
Variance	206.9	82.6	68.3	78.9	61.0