Online handwritten signature verification system based on the empirical mode decomposition

Abstract EN

The handwritten signature is a biometric method used to verify the identity of a person.

This work deals with an online handwritten signature verification system.

We model the handwritten signature by an analytical approach based on the Empirical Mode Decomposition (EMD).

The organized system is provided with a training module and a base of signatures.

The implemented evaluation protocol points out the interest of the adopted method and allows obtaining a recognition rate going up to 98 %, with a false acceptance rate (FAR) of 3.33 %, a false rejection rate of 4.67 % and an equal error rate (EER) of 1.83 %.