Face Recognition Based On Gabor Wavelet And Pca

Face recognition is very important artificial intelligence research field in image processing, pattern recognition, machine learning and computer vision during twenty years. It has great application in file management, system security, credit card verification, criminal identification, banking and customs control, human-computer interaction and other fields. This article describes the significance of face recognition research, development status and major face recognition methods.It focus on the application of classical face recognition methods such as PCA, LDA,2DPCA, 2DLDA and features of Gabor wavelet transform.In this paper, We give the features of PCA and Gabor wavelet.An improved algorithm for face recognition based on the classical 2D Gabor wavelet in feature extraction phase is proposed. We give two ways to combine the Gabor wavelet.One is the preprocessing and sample of image in face database.We use Gabor wavelet convolution to extract each image,corresponding to different directions and different scales of Gabor features. Then the dimensionality reduction and denoised technigue with PCA are applied to form the new training samples.Lastly, the nearest neighbor classifier is constructed and the vote decision strategy is used to determine the recognition result.The second is Gabor wavelet convolution of face database images to extract different directions and scales of Gabor features in each image. we use five scales, eight directional filter banks to get 40 gabor wavelet features. Then the dimensionality reduction and demised technique with PCA are applied to form the new training samples.Lastly, the nearest neighbor classifier is constructed and the vote decision strategy is used to determine the recognition result.This paper realizes face recognition based on the international standard ORL face database, Experiments show that the two face recognition algorithm have better recognition rate and accuracy and we achieve the desired results.