Signature verification using data glove with prominent sensors

Abstract

In order to improve the security level, we have presented an innovative approach to hand signature verification using data glove. Data glove is a new dimension in the field of virtual-reality environments, initially designed to satisfy the stringent requirements of modern motion capture and animation professionals. In this research, we try to shift the implementation of data glove from motion animation towards signature verification problem, making use of the offered multiple degrees of freedom for each finger and for the hand as well. Our proposed approach is tested in context of highly skilled forgeries with a large number of signature data sets and obtained a significant level of accuracy in signature verification with 2.22%, 2.5%, 2.8% and 2.37% of Equal Error Rates (EER) with 5-, 7-, 9-, and 14-sensor based data sets, respectively.