A preliminary analysis of the aging dorsal hand vein images

Abstract

This paper introduces a new research direction on the analysis of dorsal hand vein (DHV) images concerning one of the health indices, the aging of dorsal hands. In this way, a database of DHV images based on near infrared (NIR) from the old people has been collected and studied, with some image processing methods used in the biometric identification field.
There are two features proposed progressively in this paper based on the pixel gray value: the histogram feature and the
feature extracted from the mean pixel value of the vein and skin domain. Both of the features are classified with the use of
k-means clustering to get the centers of features representing the old and the young hand. In the recognition stage, the
Euclidean distance is employed to estimate whether hands of observation are old or not. The experimental results show that
the DHV images are related with some vital signs like the aging of the vein and skin. Features concerning pixel value have been verified effective in the preliminary analyzing stage of the aging DHV images.