A Vein Based Biometric Experiment and Some New Developments

The pattern of blood veins is unique, even in identical twins. The human palms have a broad and complicated vascular pattern and thus contain many kinds of features. The Vein Based Biometric (VBB) depends on measurement of the vascular pattern made by the blood vessels on the back of the hand. The VBB technology is the world´s leading and most promising biometric technology and can bring security and convenience to people in many areas. This paper firstly reviews the concepts and types of biometrics, then discusses VBB and vein image processing. According to the study of existing algorithms and techniques, a hand vein image segmentation and refinement algorithm is introduced in this section, a normalized image of the hand veins is processed by the Gaussian filtering and median filtering, in order to remove certain amount of speckle noise. By using local dynamic threshold segmentation NiBlack algorithm, a coarse vein image is produced. An image of veins lines with smaller distortion is obtained by refinement algorithm. It is also show some experimental results of VBB. The last section of this paper is about some new developments of biometrics.