Project Sponsor

eu-LISA

Project Partners

Hochschule Darmstadt (Steinbeis BISE)
NTNU (Mobai)
University of Salzburg (WaveLab Group)

Contact

Christoph Busch
Email: christoph.buschstw.de

Evaluation and improvement of eu-LISA synthetic biometric datasets

Introduction

Steinbeis Transfer Centre BISE and Mobai have been awarded a contract with eu-LISA to generate representative artificial data for fingerprints (FP) and facial images (FI), and the provisioning of biometric data quality requirements for the curation of test datasets. eu-LISA has started the implementation of the European Entry-Exit Systems’ biometric components. During its use, the European Entry-Exit System will capture and verify biometric data of travellers to the Schengen area at all border crossing points as well as by visa and law enforcement authorities. The Steinbeis Network is dedicated to the transfer of academic findings and knowledge into applications. Steinbeis Transfer Centre BISE is focused on biometrics under the directorship of Prof. Dr. Christoph Busch at Hochschule Darmstadt. Mobai AS is a spin-off company from the Norwegian University of Science and Technology (NTNU) commercializing technology developed at the Norwegian Biometrics laboratory. In this contract Mobai collaborates with experts from NTNU and Paris Lodron University of Salzburg (PLUS) to ensure expert quality within the field of face and fingerprint biometrics. The European Union Agency for the Operational Management of Large-Scale IT Systems in the Area of Freedom, Security and Justice (eu-LISA), operates large-scale IT systems for the implementation of the asylum, border management and migration policies of the EU.

Details

This contract covers the need of evaluating large scale synthetically generated biometric (fingerprint and face) databases with the intention to understand to which extent they could be used for biometric performance (accuracy) tests as well as solutions for tailoring of these datasets to make them fit for purpose. The main hypothesis to contrast is to which extent synthetic samples could replace the need for real images for biometric performance tests of the large scale IT systems. eu-LISA is responsible for the operational management of three large scale IT systems, namely Visa Information System (VIS), EURODAC and the Schengen Information System (SIS II). The Agency currently manages Eurodac, the second generation of the Schengen Information System (SIS II) and the Visa Information System (VIS). Further to these, eu-LISA is developing the Entry/Exit System (EES), the European Travel Information Authorization System (ETIAS) and the European Criminal Records Information System - Third-Country Nationals (ECRIS-TCN).

Tender Specification

Evaluation and improvement of eu-LISA synthetic biometric datasets

Background and objectives

As part of the Smart Borders program, according to Regulation (EU) 2017/2226 of the European Parliament and of the Council establishing an Entry/Exit System (EES). While the three systems currently operated and maintained by eu-LISA (VIS, SISII and EURODAC) include biometric operations based on fingerprint data, EES is a unique project due to its large size, very high performance requirements and the combined use of fingerprints and facial images on all external border crossing points of the Schengen Area. Therefore pre-go-live biometric performance testing requires changes to the old testing methods as well as new datasets that are representative enough to the future Production environment in terms of both data characteristics and size. The outcome of this project may also reduce dependency on real biometric data for future projects eu-LISA will conduct in the domain of biometrics.