The Analysis Of The Aging Dorsal Hand Vein Images

Nowadays in an era of information, biometrics has been widely applied in the field of information security and communication. There are various kinds of biometric features being used and researched including face, iris, finger-print and hand vein which attracts more and more researchers’attention and becomes a focus in the field of pattern recognition. Apart from the commonly applications in human identification and verification, some biometric features have been used in the fields beyond that.Dorsal hand vein, one of the biometrics, is a vital sign of person’s identity. However, the very biometric could be further used as a characteristic for medical diagnosis of human like aging as discussed in this thesis. Automatic recognition of aging dorsal hand vein is an interesting and striking component for a wide spectrum of applications including human-computer interfaces, medical image processing and pattern recognition to mention a few. Aging representation is an essential part in the analysis of aging dorsal hand vein recognition. It is concerned with finding distinguishable features that can differentiate the old from the young. This thesis gives a preliminary analysis on the aging representation and recognition of dorsal hand vein images. The author has built up a dorsal hand vein images database including both the old and the young. Based on the database, comprehensive analysis of the aging dorsal hand vein images has been done.Firstly, the database is set up by capturing1000dorsal hand vein images from50persons, including40old people who are over60years old; Secondly, methods used in hand vein recognition field have been employed with a regard to image pre-processing, such as noise reduction, filtering, extraction of ROI and segmentation; Thirdly, according to the observation on the database, two features are preliminarily proposed for the distinction between the old and the young:histogram and local mean value combined. At last, the two features are sent to k-mean clustering respectively for the analysis of classification and finding the centers of two clusters, which represents the old and the young. The experimental results show that aging dorsal hand vein images could be distinguished and recognized, which test and verify the feasibility of researches on aging dorsal hand vein images, and lay a foundation for further research on integrated analysis of hand dorsal vein images and medical image processing.