On-line Verification of Signatures by Fitting a Generated Aggregation Operator to Empirical Data

This paper presents a new technique for on-line signature verification. In this technique, the test signature is compared to all of the signatures in the reference set, resulting in several distance values. Then, these distance values are combined into a single value, representing the dissimilarity of the test signature to the reference set. To combine the dissimilarity values, an aggregation function from OWA operators family is constructed from training data. After combining the distance values to a single value, we compare it to a threshold to make a decision. Experimental results over a biometrical database containing skilled forgeries show the effectiveness of the proposed technique in signature verification.