Joint iris and facial recognition based on feature fusion and biomimetic pattern recognition

Abstract

Fusion biometric recognition modal contributes in two aspects. It can not only improve the biometric recognition accuracy, but also gives a comparatively safe strategy, since it is difficult for intruders to achieve multi-biometric information simultaneously, especially the iris information. In this paper, a novel biometric fusion recognition modal with iris and facial images based on biomimetic pattern recognition is proposed. The Contourlet transform (CT) and two directional two dimensional principal component analysis (2D)2PCA are used here to extract the iris feature and the facial feature respectively, and a new fusion feature vector was formed on the combination of the previous iris and facial features. Lastly, the fusion feature vector is used to construct the covering of high dimensional space using biomimetic pattern recognition method, in which the hyper-sausage neuron is adopted. Furthermore, a fixed random matrix is used here to reduce the computational complexity and improve the recognition efficiency. Experiments on the public union database show that the proposed modal can achieve the state-of-the-art recognition accuracy while keeping the enrollment process safe.