Facial feature detection is crucial for visual recognition of human faces. This paper reports the result on a study in detecting face contours from front-view ID-type pictures. A simple and effective way is proposed to detect the eye and mouth positions. Based on the relevant subimage defined in Ref.5

the location and curvature of the chin line

together with the cheek lines can be detected and the shape of the face was characterized. Especially in this paper

the eye and mouth contour sex tractions become easier and more accurate.