Identity verification by using handprint

  • Authors:
  • Hao Ying;Tan Tieniu;Sun Zhenan;Han Yufei

  • Affiliations:
  • National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, China;National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, China;National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, China;National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, China

  • Venue:
  • ICB'07 Proceedings of the 2007 international conference on Advances in Biometrics
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

In recent years, palmprint based personal identification has been extensively explored by researchers. The success of this technology has demonstrated that the inner part of palm skin is capable of distinguishing one person from another in case that proper representation is utilized. However, earlier work mainly focused on scenarios where the position and pose of hands are constrained by pegs or plates. In contrast, our purpose is to design and implement a system which is capable of recognizing an individual once he/she naturally stretches his/her hand in front of the camera. Since human hand is an articulated object, it is important to filter out geometry variations. This paper presents and compares two hand texture based personal identification methods, which are called hand-print verification in this paper to denote the idea of utilizing whole hand skin image for recognition. In one of the method, hand articulation is eliminated in a well-defined way and then the hand is treated as a whole for feature extraction, while in the other method, features are extracted in different parts of the hand, and final decision is made in a matching score level fusion manner. Experimental results for both methods are presented and compared.