Face Recognition via Deep Learning and Constraint Sparse Representation

The matching between single image and single image has attracted much attention in the current face feature matching algorithms,in order to make use of the correlation information between image sequences effectively,a face feature matching algorithm based on deep learning and constraint sparse representation was proposed.The feature extraction of face images was carried out through CNN network,and an improved sparse expression method was used to automatically select similar image sequences for feature matching,so that the correlation information between image sequences could be effectively utilized.Experimental results show that the proposed algorithm can achieve better result in LFW and AR databases,and is superior to other face feature matching methods,such as SRC,L1-norm and CRC-RLS.