Representation-based Nearest Feature Plane for Pattern Recognition

In this paper, an improved method based on nearest feature plane (NFP), called as representation-based nearest feature plane (RNFP), is proposed for biometric recognition. Borrowing the concept from the nearest neighbor plane (NNP) classifier and center-based nearest neighbor (CNN) classifier, RNFP chooses the valuable representation of the class to reduce the computational complexity of NFP. A large number of experiments on some databases are used to evaluate the proposed algorithm and the result demonstrates that the proposed method take lower computational complexity and achieve similar performance of NFP.