Face recognition evaluation of an association cortex - entorhinal cortex hippocampal formation model by successive learning

Abstract

Based on the visual information processing system in the brain, we developed an association cortex to entorhinal - hippocampal neural network model (AEH model) for learning and recollection of human faces. We evaluated the learning and recollection performance of the AEH model by computer simulation. We presented each human face to the AEH model twice. We regarded that the AEH model recognized each face correctly if the AEH model learned the face the first time and recalled it the second time, in accordance with human perception. In the experiment, ten human faces were successively presented to the AEH model. The recognition performance of the model was 100 percent correct, indicating the effectiveness of the representation and transformation algorithm of the AEH model.