An online biometric identification system based on two dimensional Fisher linear discriminant

Palm vein recognition technology is an advanced technology. It has high security and high degree of user acceptance. In this paper, a novel online biometric identification system based on palm vein is proposed. The characteristics extraction algorithm of palm vein image is two dimensional Fisher linear discriminant (2DFLD). This paper puts forward that projecting the palm vein image matrix based on 2DFLD directly, then the system calculate the Euclidean distance projection matrix, classification system based on the most recent distance. We finished the experiment on our own database. We test the proposed method and then compare the recognition performances with others typical palm vein identification methods such as SIFT, Gabor and PCA+LPP. The experimental results show that this method can be applied online for fast speed and it has advantage in recognition performance. We are now working on enhance the environment it can be apply.