The Study Of Face Recognition Methods

The biological characteristic recognition is a kind of identification technology that uses the human's special physiology or behavior characteristic, it provided a kind of high reliability, good stability approach of identity appreciation. Face detection and Recognition is very popular branch of the biological characteristic recognition. And it is also a very active subject in the fields of the computer vision and the pattern recognition.This paper has done deep research on the Face Recognition by structure a system of video image face Recognition, the main job and contribution as follows:First, The face detection is the first step of Face Recognition, this paper adopt the algorithm abstracting the haar characteristic from human face and inhuman face and then using adaboost training algorithm classifier to achieve the face diction. And do some pretreatment work that carrying out size, gradation on the face that already been detected. This paper also suggests that the wavelet transform could be used in image processing; Pick up human face's low frequency image which can be used for recognition. The experiment result show that adopting the low frequency image presents human face not only depress the image dimension, but also decrease the influence of light and t expression, obtain a higher recognition rate.Second, This paper also did detailed discussion and experiment analysis to three kinds of face recognition: PCA( principal component analyses), Fisher face method, KPAC(Kernel principal component analyses). The experiment result show that the Fisher face method get a better result than PCA by quadratic abstraction of PCA characteristic. KPCA is a kind of linearity PCA's nonlinear extend arithmetic. It takes out the main component using the nonlinear method. Describe the pertinency among several pixels. And it also could transform these questions from import space where can't be linearity classified to character space where can be linearity classified. Then get a better result.Third, This paper also suggested that the Support Vector Machines could be used in face classification and recognition. First, obtained characteristic from human face image using KPCA method. Then used Support Vector Machines to classify and recognize. This method could obtain few but extractive characteristic, little computational complexity, fine effect.Fourth, This paper also construct a human face recognition system based on static state picture and an online video human face diction and recognition system. The first one is mainly used for the static state human face database. The second one contains image login. Face detection. Human face picture pretreatment, characteristic pick-up, classify recognition, Can do face diction and recognition on the online video. It is very useful in the field of aptitude monitoring. It can be used in the laboratory as work attendance system.