Machine Learning-Based Classification for Crop-Type Mapping Using the Fusion of High-Resolution Satellite Imagery in a Semiarid Area

Table 4

Overall accuracies and Kappa coefficients of noncombined scenarios.

	SVM		ML		ANN
	OA	K	OA	K	OA	K
NDVI	85.83	0.81	84.38	0.79	81.54	0,76
VV	64.81	0.52	62.56	0.50	60.45	0,48
VH	71.35	0.61	56.96	0.44	66.90	0,55
VV/VH	51.28	0.33	45.98	0.29	44.86	0,27
Texture	58.22	0.44	50.92	0.38	54.85	0,40
VV; VH	75.43	0.67	71.15	0.60	65.62	0,56
VV; VH; texture	76.72	0.69	72.83	0.63	54.89	0,41
VV; VV/VH	72.97	0.59	66.96	0.55	58.38	0,38
VH; VV/VH	72.10	0.62	66.95	0.55	Overestimation
VV; VH; VV/VH	75.42	0.67	Overestimation		Overestimation
VV; VH; VV/VH; texture	76.45	0.68	Overestimation		55.69	0.42