A lightweight Model of Finger Vein Recognition Technology
In order to improve the ability of convolutional neural networks to recognize finger veins,a MobileNet network combining the importance of channels is proposed.Specifically,the channel importance analysis is carried out on the features extracted by Mo-bileNet,and the less important feature channels are compressed to improve the capability of network feature representation.At the same time,the network model extracted by triplet loss has the characteristics of inter-class distribution and intra-class tight-ness,which improves the discriminant ability of the network model.Experiments are carried out on MultiView-FV venous datas-et,and the results show that the method is effective.
万方数据
