A score-level fusion fingerprint indexing approach based on minutiae vicinity and minutia cylinder-code

Due to the uniqueness and permanence properties of the biometric fingerprint characteristic, large scale in border control and governmental applications such as the Visa Information System (VIS) in Europe, US-VISIT / IDENT system in the USA and the Aadhaar project in India are based on fingerprint recognition. These systems generally contain millions of fingerprint samples. In order to improve the efficiency in seeking for suitable candidate reference data in such large-scale databases, studying indexing techniques for fingerprints is desirable. In this paper, we design a new indexing method using the features extracted from minutia details including location, direction and ridge information. Then a score-level fusion indexing approach is proposed by combing this new method with the minutia cylinder-code (MCC) indexing method. The results demonstrate the improvement of the proposed approach based on experiments on several public fingerprint databases.