Table 5.

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

Net	Bands	Accuracy (%)	Precision (%)	Recall (%)	F1- score (%)
AlexNet	Delta	80.10 ± 0.75	80.21 ± 0.72	80.79 ± 0.73	80.45 ± 0.72
	Theta	80.32 ± 0.73	80.36 ± 0.71	80.21 ± 0.74	80.35 ± 0.70
	Alpha	82.40 ± 0.70	82.39 ± 0.69	82.39 ± 0.75	82.39 ± 0.69
	Beta	81.12 ± 0.77	81.15 ± 0.70	80.87 ± 0.74	81.03 ± 0.71
	Gamma	81.34 ± 0.75	81.30 ± 0.74	81.05 ± 0.75	81.15 ± 0.69
VGG-19	Delta	87.62 ± 0.75	87.55 ± 0.75	87.05 ± 0.68	87.34 ± 0.71
	Theta	88.50 ± 0.73	88.47 ± 0.75	88.10 ± 0.69	88.25 ± 0.68
	Alpha	89.95 ± 0.70	89.95 ± 0.69	89.95 ± 0.70	89.95 ± 0.69
	Beta	89.50 ± 0.70	89.47 ± 0.70	89.08 ± 0.69	89.34 ± 0.67
	Gamma	88.25 ± 0.71	88.20 ± 0.70	87.86 ± 0.68	88.10 ± 0.65
ResNet-50	Delta	97.10 ± 0.53	97.12 ± 0.65	96.85 ± 0.67	97.05 ± 0.63
	Theta	97.05 ± 0.60	97.10 ± 0.65	96.83 ± 0.65	96.92 ± 0.60
	Alpha	98.16 ± 0.54	98.16 ± 0.64	98.16 ± 0.63	98.16 ± 0.55
	Beta	97.13 ± 0.56	97.16 ± 0.63	96.95 ± 0.65	97.04 ± 0.58
	Gamma	97.15 ± 0.50	97.20 ± 0.63	96.91 ± 0.64	97.01 ± 0.59
Inception-v3	Delta	97.20 ± 0.52	97.35 ± 0.65	96.50 ± 0.66	97.14 ± 0.59
	Theta	97.35 ± 0.54	97.42 ± 0.64	97.05 ± 0.65	97.35 ± 0.58
	Alpha	97.54 ± 0.55	97.53 ± 0.65	97.53 ± 0.62	97.53 ± 0.57
	Beta	97.25 ± 0.50	97.30 ± 0.65	96.86 ± 0.64	97.17 ± 0.57
	Gamma	97.50 ± 0.48	97.57 ± 0.62	96.73 ± 0.63	97.28 ± 0.59

Highest results are showed bold