Face recognition system are welcomed in military,commerce,security,information areas because its convenience and non-invasive. After more than 30 years'development, face recognition has made great progresses especially in the past ten years. But in many application conditions there is only one training image per person, in this case most of the reported face recognition algorithms will suffer serious performance drop or even fail to work. To do special research on face recognition with single training image will be the profound significance of the face recognition application and theoretical research.In face detect, use skin color mode, accomplish human skin division and the coarse location of feature point, and after the location of eye feature points, propose a thesis that locate the center of eye with character of circle. And with the imporved canny edge detection, accomplished the accurate location of eye feature points. With Lab color model, porpose a thseis that relocation of feature points, improve the accurate of loaction.In face recognition, based with nearest-neighbour decision method, propose a mothed that get the similarity with feature vector. Whit the feature point, get the similarity with algebraic feature, but because the limitations of the location of geometric characteristic, then we should do recongnition based on gray scale informationThe result of experiment shows that the algorithm that this thesis proposed can get better recognition rate. |