Increasingly, biometric methods are being used for personal authentication or identification in scenarios such as security services, access control or internet commerce. This development has come about because the authentication system currently in widespread use only provide a limited level of security as they are based on passwords and PINs, which can be lost, stolen or hacked.

Consequently, there are a number of areas in which there is a need for improved security solutions for access authorisation. Semi-automatic or fully automatic, biometric-based systems can boost the efficiency of such solutions, especially when large volumes of data are involved. As well as improving security, system development in this area also has to ensure compliance with data protection requirements and ethical criteria. It is to this end that the project groups include researchers from the legal sphere as well as from the humanities and social sciences.

    

Completed joint projects:

GES-3D: Multi-biometric face identificatio

Funding codes  13N11998 to 13N12004

The goal of the GES-3D project was to develop a user-friendly, multi-biometric system for identifying people based on photographic or video data, using three-dimensional facial image data. From the very beginning, the research also looked at matters concerning people’s right to determine what information is available on them; data security requirements were also examined and data protection aspects were taken into consideration when developing the methods. This innovation is intended to assist law enforcement authorities by enabling them to identify individuals even if they only have a poor-quality, incomplete photograph of their face.

More information  (only available in German)

    

    

    

MARS: Mobile authentication using retina scanning

Funding codes  13N11757 to 13N11764 and 13N11997

The rise in unauthorised access to user accounts and attempts to gain unauthorised access to critical infrastructure sites poses a new challenge for many fields, including civil security. The aim of the MARS project was to devise an innovative technique for verifying users' access rights for mobile applications. Essentially the focus was on integrating retina scanners into mobile telephones. The work plan for the project includes scientific studies on health risks, data security, data protection and ergonomics. The intention is to ensure, from the outset, that the project explores all of the issues that might arise if such a method were to be used.

More information  (only available in German)

     

MisPel: Multi-biometric forensic person searches in large image and video databases

Funding codes  13N12059 to 13N12065

Nowadays, many places where security is a requirement are fitted with video cameras as standard. But the increase in footage does not necessarily mean an increase in security. In particular, cases that require video material to be used to identify criminals present significant challenges for investigators. The software and video management systems currently on sale tend to focus on efficient and elaborate visualisation methods and largely neglect the content analysis of the images. The goal of the MisPel project was to develop a software solution that helps investigators identify quickly individuals of relevance to their cases. The legal research being conducted on the project was concentrating on analysing the legal requirements that have to be met in order for data to be captured, filtered and forwarded as well as assessing the extent to which evaluated image data can be used in court. The plan was to formulate a suitable, integrated data protection and security strategy. The social science research being done was therefore designed to deliver example criteria for privacy protection plus strategies for long-term deployment scenarios.

More information  (only available in German)