Research On Methods Of Eye Location In Face Recognition

With the rapid development of information security technology, biometrics has followed rapidly. Compared with other biometrics technologies, face recognition technology is hidden, friend and convenient. It has become a hot topic both in the field of applied mathematics and high technology.In face recognition, eye location is used for the geometry normalization of a face image, and its accuracy directly affects the recognition rate. Taking both near-infrared (NIR) and visible light face databases as the research object, two effective methods of eye location are proposed in this paper which are based on the position of human face. The main contributions are summarized as follows:Firstly, according to reflection points in the pupils of NIR face images, all the NIR face images are classified into normal NIR face images and non-normal NIR face images.Secondly, for the normal NIR face images, a method of eye location which is relied on reflection points is studied, for non-normal NIR face images.and visible light face images, a method of eye location which is based on adaboost algorithm, gray and geometry feature is proposed.Thirdly, The methods proposed have been tested in the CAS-PEAL-R1database and CASIANIR database separately, and the accuracy rate are98.06%and98.86%respectively.Finally, difficult points of eye location are discussed and summarized which need to be considered in the future research of eye location.In a word, the eye location module developed in this paper has a good, stable performance and successfully applied to the face recognition system developed by the Applied Mathematics Laboratory of Northeastern University.