关键词: 人脸识别, 开放环境, Gabor滤波器, 均值, 方差

Abstract: Although progress in face recognition has been encouraged in constraint conditions, many problems still exist in unconstraint tasks. Also, it is not easy to collect a large number of face samples of each people. To solve the problems above, the task of face recognition with a small number of training images of each people is considered. A new feature extraction method is proposed based on Gabor filter. Experiments in small sample case are carried out under JAFFE and ORL dataset. The result and analysis verify the validity and robustness of proposed method.

Key words: face recognition, unconstrain tasks, Gabor filter, mean, variance