Dynamic signature verification (DSV) uses the behavioral biometrics of a handwritten signature to confirm the identity of a computer user. This thesis presents a novel stroke-based algorithm for DSV. After individual strokes are identified, a significant stroke is discriminated by the maximum correlation with respect to the reference signatures. Between each pair of signatures, the local correlation comparisons are computed between portions of the pressure and velocity signals using segment alignment by elastic matching. Experimental results were obtained for signatures from 25 volunteers over a four-month period. The result shows that when adding stroke-based features into a non-stroke feature system, the accuracy of the signature verification has been greatly improved to False Reject Rate (FRR) of 6.67% and False Accept Rate (FAR) of 13.33%. The result shows that stroke based features are important features, contain robust dynamic information, and offer greater accuracy for dynamic signature verification, in comparison to results without using non-stroke based features.
展开▼
