Year of Publication: 2012
Page Numbers: 114-122
Authors: Abdelhamid Abdesselam, Ameera Al-Busaidi
Conference Name: The International Conference on Computing, Networking and Digital Technologies (ICCNDT2012)
- Bahrain


This paper describes our recent research work on hand-based person identification. The main focus is on the use of geometrical characteristics of the hand in the identification process. Several state-of-the-art person identification methods using hand geometry were reviewed and the most commonly used geometrical features were identified; a recent research work [1] that has included most of these features was selected for implementation and evaluation. We have also tested some new geometrical features and evaluated their impact on the accuracy rate of the identification system. A set of experiments was conducted on a hand image database provided by Bogazici University, in Turkey (Bosphorus database). The results of these experiments demonstrated that the identification system based on the set of features proposed by Prasad et al is of high recognition rate, and allowed us to identify a new geometrical feature that further improve this rate.