Research of on-line handwriting signature verification algorithm based on information fusion
Zhan En-qi · Jisuanji yingyong yanjiu · 2010
In order to improve this verification effect, the paper proposed the algorithm of on-line handwriting signature verification based on information fusion.The algorithm fused three kinds of algorithms of on-line handwriting signature verification respectively based on evolution computation, neural network and discrete Frechet distance. Verified the test signature by the three kinds of algorithms to obtain its confidence level of every algorithm. Fused the last confidence level of the signature by the confidence level of every algorithm by calculating their weights, and it could be used to judge whether the test signature was genuine. The experiment results reveal the remarkably decreasing percentage of false rejection and false acceptance.