Research And Wavelet Analysis Based On Face Recognition And NMF

Face recognition is the biometric identification technology of which use visual feature information of the face to identity recognition.Compared Face recognition with other traditional biological recognition technology,Face recognition has the advantages of collection easily,with a fast and convenient,interactive friendly and so on,Face recognition has gradually accepted.The traditional ways of sub-space face recognition methods, have low processing speed and low identification rate, this paper emphasis to discuss the Wavelet analysis based non-Negative Matrix Factorization in face recognition.Face recognition system includes there basic sub-modules:pre-processing of image, face detection and localization, feature extraction and recognition. First, using the Lifting scheme Wavelet to reduce the dimensions of the image, then using NMF algorithm to feature extraction and the training image matrix decomposition of the based image matrix and sparse matrix in the form of multiplication. Finally, classify the images to be recognized. Experiment on the standardized ORL and YALE face database experiment show that the method has good recognition rate.