Nonlinear separation of signature trajectories for on-line personal authentication

Authentication of individuals is rapidly becoming an important issue. The paper proposes a new nonlinear algorithm for pen-input on-line signature verification incorporating pen-position, pen-pressure and pen-inclination trajectories. A preliminary experiment was performed on a database consisting of 1849 genuine signatures and 3174 skilled forgery signatures from fourteen individuals. False acceptance rates and false rejection rates of less than 2% were obtained. Since no fine tuning was done, this preliminary result looks very promising.