Practical face recognition system using eigenface algorithm

A method taking multi-samples as sub-modes and grouping face modes into par tial intersection ones was proposed to reduce computation and improve syste m extension property. In combination, the sum rule based on Bayesian theory was used. The face recognition experiments with the ORL and AR face databa ses showed that the eigenface algorithm using multi-samples reached a high recognition rate and a reasonable time cost. Grouping face modes makes trai ning become a distributed computation job which reduces time cost for train ing and brings the convenience of system extension when new face modes are to be added.