Personal identification based on finger vein and contour point clouds matching
Zhaochun Ma, Liyong Fang, Jianhua Duan, Song Lian Xie, Ziqian Wang · 2016
Finger vein recognition has attracted increasing attention in biometric identification field. To solve the problems of lack of depth information in recognition based on 2D image and no enough obvious features used for recognition based on 3D point cloud, a novel approach of identifying individuals using 3D point clouds matching of finger vein and contour is proposed in this paper. A pair of vein images of a finger are captured under Near Infrared(NIR) light using our binocular vision device. The edge and vein contours of finger are extracted as features used for describe finger vein and then 3D point cloud of finger vein and contour is reconstructed though binocular vision technique. At last, personal identification based on matching results between template and reconstructed 3D point cloud using Iterative Closest Point(ICP) algorithm can be achieved. The experimental results show that proposed method can generate more discriminative features to represent 3D model of finger vein and contour so that the accuracy of matching is enhanced.