Automated identity verification based on hand shapes

We present a method for identity verification based on matching of hand shapes. Our method first represents the shapes of hands by sets of ordered points. Next we match the two sets of points using., our developed point matching algorithm based on deterministic annealing and compute matching parameters. Considering the research of hand shapes verification usually work in practical cases of limited or small samples, we develop support vector machine classifier for verification. We also present our preliminary verification results based on a small hand shapes database. The results show the effectiveness of our method.