Abstract: It's a key problem to abtain appropriately low-dimensioned face features in the face recognition.A face recognition method bas;ed on wavelet packet and Principal Component Analysis (PCA) is presented.Wavelet packet transform can resave the main information and details in different level.Firstly, Wavelet packet transform decomposed an image into low frequency hand and high frequency hand.Than, chooses the best-basis.Lastly, recognizes with the image based on PCA.The experimental results on Yale data-bases demonstrate the high efficiency of the algorithm in runtime and correct localization rate.