Algorithm of accurate eye detection of frontal facial images
正面人脸的眼睛精确定位算法

Using the scatter matrix that is related to the gray-level of irises, the algorithm first extracted intensity valleys as the candidates for the irises. Next, for each pair of candidates, the algorithm computed a cost using Hough transform and separability filter to measure the fit of the pair of candidates to the image. Experiments were carried out by using all faces without spectacles in the face database of the SJTUFace and BioID, the success rates of the proposed algorithm were 96.5% and 91.36% respectively.