中文摘要：

鉴于人脸识别面临光照、表情和遮挡等因素的影响,提出了一种在分数阶傅里叶变换域稀疏表示的人脸识别。基于分数阶傅里叶变换对光照、表情的鲁棒性,已在图像处理领域得到应用。FRFT幅度随阶次的变换呈现压缩性,而SPCA提取其主要信息,且分为主要信息域和次要信息域,融合两者的互补信息组成混合幅度特征,然后融合混合幅度特征、实部特征和虚部特征,最后融合不同阶次下FRFT域特征。此外提出基于贪婪算法的分数阶阶次选择算法和基于Fisherfaces的权重方法。ORL和AR人脸数据库上识别率分别达到了96.5%和97.6%,充分证明了该算法对人脸识别的有效性。

英文摘要：

Face recognition systems suffers from illumination,expression and occlusion and so on. This paper presented a novel discrete fractional Fourier features method based on sparse principal component analysis（ SPCA） for face recognition. It used the fractional Fourier transform（ FRFT） to image processing with its robust to illumination and expression. Specially,it handled the magnitude of FRFT,whose energy displayed constringent characteristic,by SPCA to further divide into the main energy of magnitude part（ MMP） and the remaining energy of magnitude part（ RMP）,which combined into the hybrid magnitude part（ HMP） to fuse complementary features. Then for fractional Fourier features with individual transform order,the hybrid fractional Fourier features（ HFFF） formed and consisted of three fractional Fourier features： HMP,real part（ RP） and imaginary part（ IP）. Finally,it fused the HFFF generated using three fractional Fourier features with different transform orders by means of the weighted summation rule-the decision level fusion-to derive the MOFF for face recognition. In addition,it introduced the greedy search to select the transform order of the FRFT. Experimental results of MOFF on the ORL（ 96. 5%） and AR（ 97. 6%） databases verify the effectiveness of the results by using these new modifications.