DISCRIMINATION OF FACIAL REGIONS BASED ON DYNAMIC GRIDS OF POINT REPRESENTATIONS

The application of an elastic graph-matching approach to discriminate facial image regions is presented. In contrast to the dynamic link arc hitecture introduced by the Malsburg group, our application is not an identification task but a classification task. Therefore, our approach differs in several important aspects: (1) the choice of the filter se t, (2) the selection of the positions of the nodes of the graph to rep resent the characteristic image information, (3) the generation of a r epresentative reference pattern needed for the calculation of the clas sifications, and (4) a new two-step graph-matching approach based on t he simulated annealing technique. The approach was tested on facial re gions taking the eye region as an example target. A classification per formance for the verification of eye regions of more than 93% was achi eved.