Real-Time Gender Recognition for Juvenile and Adult Faces

Table 4

The optimum point (OP) of receiver operating characteristic curve (ROC) and area under the curve (AUC) for the proposed feature engineering model (Gabor-meanDWT); Protocol: five cross validation (K5); Dataset: FERET; Classifier: SVM-RBF, SVM-Linear, Naive Bayes, and logistic regression(LR), respectively.

Classifier	Gabor-meanDWT
	AUC	Optimum Point (FPR, TPR)
SVM (RBF)	0.998	0.012, 0.970 7
SVM (Linear)	1	0,1
LR	1	0,1
Naive Bayes	0.767 0	0.588, 0.946