Cartesian and polar transformation-based cancelable fingerprint template

Fingerprint authentication systems suffer from security and privacy issues since it needs to store the fingerprint template in the database. Once this data is compromised, the effect is forever as it can not be revoked. In this paper, we propose a local feature-based cancelable fingerprint template scheme. The experimental result which is generated by using a public database, FVC2002DB2, shows that it has a low error level and also meets the security and privacy requirements, such as revocability and diversity.