In this paper, we present an efficient face detection algorithm for locating human faces in the complex scenes.
The algorithm models the distribution of human skin color in the normalized rg color space and finds various face candidate regions. Thresholding with predetermined threshold levels selects face candidate regions. For each of these face candidate regions, the Canny edge operator is used to find edge regions. For each edge region, we use an ellipse detection algorithm that is similar to the Hough transform to refine the candidate region. Finally, if a substantial number of the facial features (eye, mouth) are found successfully in the candidate region, we determine the face candidate region as a face region.
We show empirically that the proposed algorithm can find the face region very well in the complex scenes.