Skin color and feature-based segmentation for face localization
Juan Jose de Dios · Optical Engineering · 2007
This paper presents a combination of a color- and a morphological-based method for face localization in color images. The skin-color segmentation technique is first applied using a color space, YCgCr, specifically proposed for face detection, in order to obtain the skin regions inside the image. Then, a combination of morphological operators and algorithms is used for completing the segmentation masks. In a second phase, the location of the facial features, based on the elliptical shape of the face, is performed for finding the face candidates. Therefore, the definition of a simple face model based on the best-fit ellipse determined by the position of the eyes and mouth is proposed, and its application to the segmentation mask yields the localization of the face. Finally, the segmentation mask based on this ellipse is used for detecting the face in the image under test.