EU publishes biometric quality requirements for the Entry/Exit System
The NFIQ 2.0 algorithm plays a key role to improve the quality assessment of fingerprints in the EES
Good quality of a fingerprint image is a prerequisite to assure a person can be reliably identified based on their fingerprint. This is especially important in government applications, such as the issuance and verification of electronic passports or visa.
For the quality assessment of fingerprints, the Fingerprint Image Quality (NFIQ) algorithm of the US National Institute of Standards and Technology (NIST) - initially developed in 2004 - has become established worldwide. As biometric technology evolved, so did the NFIQ standard to meet the requirements for secure biometrics of today as well as tomorrow.
Since 2017 the new version 2.0 of the NFIQ algorithm, which is based on artificial intelligence, has been available. It is an essential foundation for the security and reliability of modern biometric systems that use fingerprint verification. NFIQ 2.0 is now also set as the standard for fingerprint quality in the EU Entry/Exit System¹, which will be introduced in 2021.
On behalf of the German Federal Office for Information Security (BSI), secunet experts were instrumental in optimising and adapting the NFIQ 2.0 algorithm to meet the latest requirements for biometric applications. In collaboration with NIST and additional partners, secunet -on behalf of BSI- developed the concept for the improved method to assess the quality of fingerprints. Further, secunet provided a modular software framework that allowed training the NFIQ 2.0 algorithm based on the concept of machine learning, which resulted into the algorithm now set for the EU Entry/Exit system.
¹According to the Official Journal of the European Union No. L57: “COMMISSION IMPLEMENTING DECISION (EU) 2019/329 of 25 February 2019 laying down the specifications for the quality, resolution and use of fingerprints and facial image for biometric verification and identification in the Entry/Exit System (EES)”