Research On Face Recognition Technology

The automatic recognition of human face is one of the most challenging subjects in the fields of pattern recognition and computer vision. It is the aims of many scientists who work in computer science field to enable computer have human's intelligence and remember and recognize person's face just like what human does. With the development of society and the improvement of science and technology, it is urgently needed for convenient and reliable automatic status discrimination. So face recognition resumes being the highlight of machine intelligence research field. The significance of this recognition is not only to impulse the theory and application of image operation, pattern recognition, but also to appease some practical needs such as identity confirmation and the search of basic contents. Due to it's particularity, it can promote cognition science, psychology, physiology and some interrelated subjects.Learned many essays, research papers concerning human face detection and face recognition of the domestic and international in recent years, some problems about face detection and face recognition are analyzed. Aim at the two important aspect of building an automatic face recognition system, face detection and face recognition, we do some researches deeply. Experiments indicate that the methods of face detection and face recognition in this paper considered reasonable, which have certain theory value and practical value. The research work of this paper includes the following several respects mainly:1, Introduce the method of dubiety excluding to the traditional face detection algorithm based on the model of faces' color, which have improved the accuracy rate of face detection greatly through this method. Experiments on the pictures of personal face indicate that there is very high practical value in this method.2, Proposed a feature extraction approach base on local geometrical feature. As the traditional algorithm can't extract facial feature while there are shadows, use a new improved method to solve this great problem. Experiments show this method can extract facial features of people effectively.3, Adopt a projective method instead of the traditional geometrical method when extract the feature of eyebrow. Experiments show this method can keep the mostfeatures of the eyebrow.4, Proceed with the statistical facial features of people, introduce the PCA(Principal Components Analysis ) in the face recognition based on the theory of sub-space of Hilbert.5, Adopt a method of order scoring in the feature matching, improve the rate of facial recognition.