In order to improve the problem that the traditional local binary pattern（ LBP） of finger vein recognition rate is low

a finger vein identification method is proposed based on multi-block mean neighbors based binary pattern（ MMNBP）. LBP method is improved by proposing neighbor-based binary pattern（ NBP）. Finger vein image is evenly divided and extracting the features of all sub-block mean values by NBP

so as to form the MMNBP method. Match images by Hamming distance. Experiment databases are from the foreign and the domestic

and using several typical methods of identification for comparison experiments. The experimental results show that the method can get the lowest equal error rate are 2. 461 1 % and 0. 313 7 %

it is proved that the method can further improve the robustness of identity recognition

which has stability and effectiveness.