Finger-Vein Image Enhancement Based on Orientation Field

Abstract

Finger-vein image enhancement is of great importance for finger-vein recognition since the quality of the finger-vein images always is very poor in practice. In this paper, a new method based on orientation field is proposed for reliable venous region enhancement. First, a coarse vein-width variation field (CVWVF) is adaptively estimated by the curvatures of the cross-sectional profiles in a finger-vein image. Second, a line filter transform (LFT) based on a line model with CVWVF constraint is computed for a primary orientation field (POF) generation in a finger-vein image. Third, to refine POF, a curve model with CVWVF constraint is used for implementing a curve filter transform (CFT). By CFT, the venous regions can be enhanced reliably in a finger-vein image. Finally, experimental results show that the proposed method has a good performance in finger-vein image enhancement.