Advancing face detection efficiency: Utilizing classification networks for lowering false positive incidences