Table 4.

Evaluation metrics on the validation set.

Model	Class Name	Precision	Recall	F1-Score	Accuracy
Inception V3	Not drowsy	0.9852 ± 0.003	0.9888 ± 0.003	0.9870 ± 0.002	0.9870 ± 0.002
	Drowsy	0.9888 ± 0.003	0.9851 ± 0.003	0.9869 ± 0.002
VGG16	Not drowsy	0.9865 ± 0.004	0.9849 ± 0.004	0.9857 ± 0.002	0.9857 ± 0.002
	Drowsy	0.9849 ± 0.004	0.9865 ± 0.004	0.9857 ± 0.002
ResNet50V2	Not drowsy	0.9926 ± 0.003	0.9953 ± 0.003	0.9939 ± 0.002	0.9939 ± 0.002
	Drowsy	0.9953 ± 0.003	0.9926 ± 0.003	0.9939 ± 0.002
Proposed DD-AI	Not drowsy	0.9980 ± 0.000	0.9988 ± 0.001	0.9984 ± 0.001	0.9984 ± 0.001
	Drowsy	0.9988 ± 0.001	0.9980 ± 0.000	0.9984 ± 0.001