Eye detection based on multi-angle template matching

An eye detection algorithm from facial image based on multi-angle template is described in this paper. First, we get the face region using skin-color model, then we extract the edge of face region. After that, we can obtain the eye candidate regions with morphological opening and closing operations and symmetrical characteristics of the eyes. At last, we can accurately locate the position of eyes by multi-angle template matching. The experimental results show the effectiveness and the robustness of this algorithm.