Table 4.

Classification results for proposed method using different pre-trained CNNs and PDC images on DEAP database in all frequency bands while using tenfold Cross-Validation method

Net	Bands	Accuracy (%)	Precision (%)	Recall (%)	F1- score (%)
AlexNet	Delta	75.23 ± 0.46	75.24 ± 0.53	75.05 ± 0.54	75.20 ± 0.52
	Theta	75.55 ± 0.58	75.56 ± 0.59	75.64 ± 0.56	75.52 ± 0.56
	Alpha	77.12 ± 0.53	77.13 ± 0.55	77.15 ± 0.55	77.08 ± 0.45
	Beta	76.48 ± 0.62	76.15 ± 0.47	76.06 ± 0.43	76.44 ± 0.52
	Gamma	76.14 ± 0.48	76.18 ± 0.53	76.76 ± 0.48	76.12 ± 0.55
VGG-19	Delta	83.39 ± 0.52	83.35 ± 0.54	83.41 ± 0.55	83.35 ± 0.53
	Theta	83.76 ± 0.47	83.75 ± 0.49	83.78 ± 0.62	83.70 ± 0.52
	Alpha	85.42 ± 0.65	85.41 ± 0.56	85.45 ± 0.55	85.38 ± 0.56
	Beta	84.44 ± 0.48	84.42 ± 0.45	84.45 ± 0.57	84.42 ± 0.48
	Gamma	84.15 ± 0.53	84.14 ± 0.48	84.16 ± 0.43	84.12 ± 0.55
ResNet-50	Delta	91.20 ± 0.51	91.22 ± 0.43	91.26 ± 0.46	91.16 ± 0.51
	Theta	91.52 ± 0.60	91.55 ± 0.52	91.54 ± 0.56	91.50 ± 0.56
	Alpha	95.73 ± 0.56	95.70 ± 0.54	95.77 ± 0.52	95.69 ± 0.55
	Beta	93.36 ± 0.56	93.33 ± 0.54	93.42 ± 0.47	93.29 ± 0.62
	Gamma	93.24 ± 0.53	93.22 ± 0.45	93.28 ± 0.56	93.21 ± 0.54
Inception-v3	Delta	91.25 ± 0.58	91.24 ± 0.75	91.27 ± 0.48	91.21 ± 0.63
	Theta	91.44 ± 0.49	91.41 ± 0.56	91.47 ± 0.57	91.42 ± 0.56
	Alpha	94.12 ± 0.52	94.10 ± 0.47	94.14 ± 0.48	94.07 ± 0.65
	Beta	92.56 ± 0.56	92.52 ± 0.54	92.59 ± 0.53	92.54 ± 0.46
	Gamma	92.21 ± 0.54	92.18 ± 0.55	92.23 ± 0.55	92.20 ± 0.48

Highest results are showed bold