Biometric verification system based on hand geometry

Abstract

Biometric systems are widely used in medium and low security applications. Verification systems based on the geometry of the hand utilize some geometrical characteristics of the hand including measurements of fingers, shape of the palm, etc. In this work, we have developed an unconstrained and contact-based hand geometry verification system, using a combination of length and width of fingers. New measurements at different points of fingers were introduced in this paper to improve the performance of the recognition of persons. A total of 135 hand images were enrolled in this study. The Euclidean distance was used as a similarity function for different values of threshold. The proposed method was compared to state-of-the-art approaches. The results obtained reveal the high performance of the proposed approach and outperformed the existing methods with an accuracy of Acc = 98.67%.