An Effective L0 - SVM Classifier For Face Recognition Based on Haar Feature

摘要

Face recognition is an important research topic in pattern recognition, and in which, it is a striking direction that how to extract the useful features to express face. In this paper, we present a technique for face recognition by L0 -SVM classifier based on Haar features. Firstly, a mass of Haar features are produced by different kinds of Haar template. Then basing on the Haar features and according to the DC algorithm, L0-SVM classifier is constructed in order to enhance computational and time efficiency, as well as its validity is proved in theory. Finally, experimental results on databases show that the method can effectively improve the recognition rate of the face with a small scale of samples.