Research on information fusion algorithm of on-line handwriting signature verification

The information fusion algorithm of on-line handwriting signature verification is proposed in order to improve the verification effect. The fused algorithm fuses three kinds of algorithms of on-line handwriting signature verification, respectively based on the evolution computation, the neural network and the discrete Fre¿chet distance. The test signature is verified, by each of the three kinds of algorithms, to obtain its confidence level of every algorithm. The final confidence level of the signature is fused by the confidence level of every algorithm through their weights calculation, and it can be used to judge whether the test signature is genuine. The experiment results reveal the remarkably decreasing percentage of false rejection and false acceptance.