Finger vein recognition based on Hessian-Affine operator

Finger vein recognition has become a popular biometric technology due to its security and convenience. However, quality degradation of images, especially the arbitrariness of finger placement during acquisition that makes venous areas scaled, translated and rotated, often seriously decrease the rate of the recognition. To solve the problem, a finger vein recognition method based on Hessian-Affine operator is proposed in this paper. First, the region of interest (ROI) is extracted from the raw image followed by a gray scale unevenness correction for image enhancement. All affine invariant feature points extracted by Hessian-Affine operator and described by SIFT descriptor are subject to feature matching. Experiments on MLA data set published by Shandong University and its expansion set show that the Hessian-Affine feature is robust to the translation, scale, rotation and illumination changes of the vein region that helps vein recognition.