A palm vein feature extraction method based on affine invariant

Palm vein recognition is a new biometric identification technology. The horizontal rotation, translation and tilting of palm vein image greatly affect recognition rate. To solve the above problems, this paper proposed a recognition method for palm vein based on affine geometric properties. Firstly, the palm area of the palm vein image is obtained through image preprocessing. Secondly, a series of centroids of palm and segments are extracted. Feature vectors are constructed with the area radio of the triangles which are formed with centroids. Finally, Euclidean distance is used as the matching criteria. Experimental results show that the proposed method can obtain high recognition ratio and is robust to rotation and tilt of vein image.