A Sensor-Based Approach for Dynamic Signature Verification using Data Glove

摘要

Data glove is a new dimension in the field of virtual reality environments, initiallydesigned to satisfy the stringent requirements of modern motion capture andanimation professionals. In this paper we try to shift the implementation of dataglove from motion animation towards signature verification problem, making useof the offered multiple degrees of freedom for each finger and for the hand aswell. The proposed technique is based on the Singular Value Decomposition(SVD) in finding r singular vectors sensing the maximal energy of glove datamatrix A, called principal subspace, and thus account for most of the variation inthe original data, so the effective dimensionality of the data can be reduced.Having identified data glove signature through its r-th principal subspace, theauthenticity is then can be obtained by calculating the angles between thedifferent subspaces. The SVD-signature verification technique is tested withlarge number of authentic and forgery signatures and shows remarkable level ofaccuracy in finding the similarities between genuine samples as well as thedifferences between genuine-forgery trials.