Face recognition based on feature weighted and performance comparison
基于特征加权的人脸识别

Current face recognition methods usually do not consider the effects of different facial features. After the roles of different facial features in the course of face recognition were researched, face images were preprocessed by three feature weighted methods respectively, then were recognized by popular face recognition methods of association memory, principal component analysis and Fisher linear discriminant analysis. The experiments on ORL face database show that feature weighted methods are effective and general to face recognition.