Continuous versus exclusive classification for fingerprint retrieval

This work addresses the problem of fingerprint retrieval in a large database. Traditional approaches adopt exclusive classification of fingerprints; the paper shows that a continuous classification can improve the performance of fingerprint retrieval tasks significantly. The proposed approach is based on the extraction of numerical vectors from the directional images of the fingerprints; the retrieval is thus performed in a multidimensional space by using similarity criteria. © 1997 Elsevier Science B.V.