Hybrid Fingerprint Recognition Using Minutiae and Shape

This chapter presents a novel minutiae matching method that uses a thin-plate spline model to describe elastic distortions in fingerprints. The thin-plate spline model is estimated using a local and a global matching stage. After registration of the fingerprints according to the estimated model, the number of matching minutiae can be counted using very tight tolerance boxes. For deformed fingerprints, the algorithm gives considerably higher matching scores compared to rigid matching algorithms, while only taking 100 ms on a 1GHz P-III machine. The additional step of shape matching after elimination of the elastic deformations helps to reduce the error rates further. Furthermore, it is shown that the observed deformations are different from those described by theoretical models proposed in the literature. This chapter is an extension of earlier work that is reported in [1].