Based On Color Information Of Face Detection And Eye Positioning

Face recognition technology, as an important field of pattern recognition and computer vision research, is particularly significant to applications such as frontier defense security, video monitoring, and identity authentication etc. Face detection is original from face recognition, which is the key section of face recognition system. In recent years, face detection has been attached importance to by researchers as an absolute task with the development of electronic commerce. Face detection is the premise of rapid and accurate face recognition, its purpose is detecting the face from a picture background.In color images, the skin-color is the important information of face. It is stable, and independent of detail changes. The skin-color has a good character of clustering in chromatic space. According to this character, much background factor can be discarded quickly.In this dissertation, we simply introduce the concept, application and the present research condition of face detection firstly,and then analyze several commonly color space, selecting the color of a good clustering YCbCr color space to establish color model, using this model, the complexion is segmented and two-value processed. On account of noise's existence, we can get backup face areas using binarization image filtering based on mathematical morphology. In this paper, we use the adjoint method of geometric characters (acreage, aspect ratio, centre) and the Euler number to detect the face region effectively.In this paper, the important characteristic of the face organs– eyes is studied. Firstly, human eyes are located using the geometric rules roughly. Secondly, face contour is approximate ellipse, and in general to meet the geometric distribution of "three chambers five eyes", the location of eyes has a certain geometric relationship in the face region. Then the template matching method is used to locate eyes precisely.In this paper, we use Matlab to program. A coarse-to-fine search strategy has been used in face detection and eyes location. In this program, the advantages of different algorithms are complemented for each other, and system accuracy is improved.