Face recognition from single sample per class based on Gabor filtering
基于Gabor变换的每类单个训练样本人脸识别研究*

Based on high dimensional biomimetic information theory,this paper proposed a novel training samples covering algorithms using minimum spanning trees.By Gabor filtering,gained 40 Gabor sub-images,and then recognized these sub-images by HSN model.Made decision fusion by voting strategy for these 40 sub-images recognition result.Experiments performed on the YaleB_Extend and AR databases show the provided method can achieve a better performance than the PCA_NN when only using a single sample per class training...