Research Of Facial Recognition System Based On Image

Identity recognition has been paid more attention to. in addition to higher accuracy, kindness and directness of the recognition system should be considered, at last, it is worth considering whether psychologic obstacle will be brought on the people recognized, so face is chosen to become the recognized object, then a recognition system based on facial image is established, and theoretical design and model of this system is brought out. The checking algorithms of facial recognition have been mainly researched by experiments. l)to extract complexion , facial image is transformed into HSI color space, then face location is found out. 2) through project method based on general knowledge, the location of eyebrow, eye, nose, lip and chin is found out. 3) through template and active contour, exact character of facial features is extracted, for eye character, a proper template is brought out. 4) similitude of two face image is calculated out by means of distance formulate, and proper threshould is selected to judge whether the two face images belong to one person.Experiment results show: Dface location is found out on basis of colorinformation, speed is quicker, algorithms is simpler, and location is nicety. 2)in edge image, the amount of isolating point is reduced, the correctness of following location of eyebrow, eye, nose, lip and chin is insured. 3) important character points are extracted through new template, peed is quicker, algorithms is simpler. 4)this system is more exactly and efficiently.