Gabor-Based Kernel Independent Component Analysis for Face Recognition

In this paper, a new Gabor-based Kernel Independent Component Analysis (GKICA) method for face recognition is presented. This method first derives a Gabor feature vector from a set of down-sampled Gabor wavelet representations of face images, then reduces the dimensionality of the vector by means of kernel principal component analysis, and finally defines the independent Gabor kernel features based on the Independent Component Analysis (ICA). Experiments are performed to test the proposed algorithm on ORL dataset and Yale dataset. Results show that our new algorithm achieves higher recognition rates than ICA and Kernel Independent Component Analysis (KICA), and costs less time than Gable-based ICA (GICA).