Font Size: a A A

Researches Of Face Recognition Methods Based On SVD And Neural Networks

Posted on:2006-08-29Degree:MasterType:Thesis
Country:ChinaCandidate:H X JiaFull Text:PDF
GTID:2168360155958073Subject:Electrical theory and new technology
Abstract/Summary:PDF Full Text Request
Biometrics, because of using the proper living creature characteristic of human body, is the totally brand -new technique different from traditional personal identification method. Because it has the better safety, dependable with the usefulness, more and more people thoughtful of. In all kinds of methods, face recognition technique, one for personal identification, is accepted most easily. The automatic recognition of human faces is an active subject in the area of computer vision and pattern recognition over the past few years, which has a wide range of potential applications in the areas of public security, identification of certificate, entrance control and video surveillance.The development and mainly methods of face recognition technique are firstly introduced in this paper. Then the problems of facial features localization, feature extraction and classification are discussed. Finally, we put forward a method for the recognition of full-face and immobile human faces images, the work including:(1) Face detection: For the particularity of the face image of this paper, That is the picture have no big carriage variety, and the background is not complicated, We put forward a facial features localization method based on integration projection theory and transcendent knowledge of human face. Firstly, we can get face breadth through vertical projection method. Then we can cut roughly the face picture into a rectangle outline by using transcendent knowledge of human face. Finally, through a series of standardization, we can get the whole face picture, which is located accurately.(2) Feature extraction: For the singular value can reflect the character of matrix effectively, We extract face feature based on SVD (Singular Value Decomposition) method. Firstly, we can get singular value characteristic vector by SVD theorem, then through a series of singular value transformation, which include singular value declines dimension, singular value vector standardizes, singular value vector compositor, finally, we can get singular value...
Keywords/Search Tags:Biometrics, Face Detection, Integration Projection, SVD (Singular Value Decomposition), Face Recognition
PDF Full Text Request
Related items