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

Table 5

Overall accuracy and Kappa indices of combined scenarios.

	SVM		ML		ANN
	OA	K	OA	K	OA	K
NDVI; VV	82.93	0.77	81.24	0.74	59.98	0.45
NDVI; VH	86.10	0.81	84.46	0.78	72.18	0.62
NDVI; VH; texture	87.22	0.82	85.68	0.80	51.89	0.38
NDVI; VV/VH	82.33	0.80	81.00	0.74	58.11	0.43
NDVI; VV; VH	83.78	0.78	82.83	0.76	Overestimation
NDVI; VV; VV/VH	83.37	0.77	83.39	0.76	Overestimation
NDVI; VH; VV/VH	84.60	0.79	82.39	0.75	Overestimation
NDVI; VH; VV/VH; texture	85.52	0.80	84.36	0.78	55.09	0.42
NDVI; VV; VH; VV/VH	83.83	0.78	Overestimation		Overestimation
NDVI; VV; VH; VV/VH; texture	85.20	0.80	Overestimation		55.59	0.42