Digital Mapping of Soil Drainage Classes Using Multitemporal RADARSAT-1 and ASTER Images and Soil Survey Data

Table 4

Classification accuracy (%)* of land use types using supervised maximum likelihood classification applied on selected ASTER image attributes.

Land use type		Observed classes
(LUT)		N1	N2	C	P	F	H	W
Predicted classes	N1	81.6	4.9	1.5	11.2	1.5	0.3	0.2
	N2	13.3	93.2	0.1	4.2	0.1	0.0	0.0
	C	1.6	0.2	84.0	12.6	6.2	0.8	0.6
	P	2.5	1.7	2.0	85.4	2.8	2.9	0.0
	F	0.4	0.0	5.2	3.0	72.4	3.1	6.7
	H	0.7	0.0	6.3	3.6	17.0	92.9	1.9
	W	0.0	0.0	0.0	0.0	0.0	0.0	90.6

*Overall classification accuracy = 86%; Kappa coefficient = 0.50.