Skip to Main Content
The applicability of universal background models as a score normalization technique is studied for the case of dynamic signature verification. This technique is commonly used in speaker verification systems. Background models are tested in two different systems based on global features: one based on Parzen windows and another based on adapted Gaussian mixture models. Experiments are carried out in the large MCYT database (16,500 signatures from 330 users) revealing a significant improvement in the overall system performance, specially in the casual impostor scenario.