Variation of grip force profile during signature writing

To examine the possibility of differentiating between individuals based on the total grip force profile generated while writing a signature, this study investigated the variability of features derived from grip force profiles of twenty adult participants writing the same signature over multiple sessions. Using an instrumented writing utensil and a digitizing tablet, each participant provided 600 samples of a well-practiced bogus signature over a period of 10 days. Using a linear discriminant analysis classifier with a combination of temporal, spectral and information-theoretic features, discrimination among participants was possible with an average misclassification rate of 5.8%. This linear classifier outperformed nonparametric and nonlinear classifier alternatives that were tested in this study. These results suggest the existence of a unique kinetic profile for each writer even when writing the same signature and despite some variation within each writer. The results of this study also highlight the potential of using grip kinetics as a biometric measure for signature verification applications.