Face Recognition Based on Manifold Ranking and PCA

In this paper, we propose a novel framework for face recognition based on manifold ranking algorithm and PCA algorithm. First, manifold ranking algorithm is used to explore the relationship among all the data points in the feature space, and then constructs a weighted graph with respect to the intrinsic manifold structure collectively revealed by all the space data. So the images can be measured and ranked respect to the similarity of image data. The front r images of one category in the retrieval result are selected because of no accord between such manifold structure and the semantic concept belong to the same image. Finally, the PCA algorithm is applied to face recognition on the selected images and the sample face image. The experimental results from face images data illustrate the validity of our method.