A New Method of Fusion of ICA and LDA for Face Recognition
一种基于ICA和LDA组合的人脸识别新方法

Feature extraction is one of the hot topics in the field of pattern recognition.Independent Component Analy- sis(ICA)and Linear Discriminant Analysis(LDA)are two effective methods of feature extraction.ICA can extract features from high order and LDA can extract features which are useful to classify.However,performance of ICA method combined with LDA is reported as lower than ICA method previously.In this paper,we point out the weakness of the previous methods and a new method of feature extraction based on ICA and LDA with conception of null space is proposed.Experiment results carried on face databases demonstrate the effectiveness of the proposed method,