Face Recognition Based on Gabor Features and Unit-Linking PCNN

In this paper, a novel method is proposed for face recognition based on combined Gabor features and unit-linking pulse coupled neural network (PCNN) time signature. In this approach, a probe face is first divided into a 2times2 block then the mean and standard deviation of the Gabor sub-face image are extracted which includes 40-dimension features in each block. The PCNN time signature of a face image is combined with the Gabor features as the recognition features, which is classified with Euclidean distance or support vector machine. An extensive experimental investigation is conducted using AT&T face database covering face recognition under controlled/ideal conditions, different illumination conditions and different facial expressions. The recognition ratio is 87% when only Gabor feature, including three exemplar images per person are available. The recognition ratio is 95.25% when Gabor feature is combined with the unit-linking PCNN time signature while incurring little increase to the processing time since the dimension of the unit-linking PCNN time signature is low.