The Research On Face Recognition And Algorithm Of Eye Location

The technology of face recognition is one of the most interesting research directions for its wide application. The human face is a complex pattern. Though it is easy for a child to recognize a human face, it is difficult for a computer to recognize a human face automatically.A general system efface recognition includes: pretreatment, feature extraction , sample study and sample recognize. As the most important step, feature extraction will affect the rate of face recognition directly, and most of the methods of feature extraction depend on eye location. And a new method of eye location is proposed in this paper that is based on the character that eyes have higher degree of complexity compared to whole face.In this paper, first, several methods of pretreatment to face images are introduced; then, some typical and effective methods of feature extraction are discussed, such as methods of geometry feature extraction, methods of algebra feature extraction, methods based on neural network and so on; and then, some methods of eye location in common use are given; last, a new method of eye location and its experimental data are proposed, the method is based on the character that eyes have higher degree of complexity .It is proved by experiments that the correct rate of eye location is 100 percent by new method based on experimental data. What's more, the new method is easy to realize and is quicker in velocity, so it will meet the demand of real time application well.