The University of Kentucky is testing biometric facial authentication for access control via a pilot with students at a residence hall. It is an interesting example of how biometrics are likely to complement other forms of ID, as Kentucky is using it alongside existing cards and mobile credentials.

The pilot started earlier this year at Woodland Glen IV, a residence hall on the Lexington campus. The project is being conducted in partnership with the UK Police Department and biometric technology provider Alcatraz.

According to the university, the goal is to evaluate whether biometrics can improve both security and convenience at exterior residence hall entrances.

UK views residence hall security as a layered process that starts with locks on individual rooms and extends to controlled building and floor access, front-desk procedures, cameras, and exterior security patrols.

Students participating in the pilot use their face to authenticate at a building entrance rather than presenting a plastic Wildcard or mobile credential. Participation is voluntary, and existing credentials remain operational for students who do not participate or if facial authentication is unsuccessful.

Addressing the tailgating problem

While hands-free entry provides a convenience benefit, campus security officials are also evaluating the technology's ability to address tailgating.

Tailgating occurs when an unauthorized individual enters a controlled building by following an authorized person through an open door without presenting a credential.

The system can detect tailgating in real time and share this information with the existing physical access control infrastructure.

UK Police Chief Joe Monroe discussed the university's interest in facial authentication earlier this year as part of a broader conversation about residence hall security.

He described residence hall security as a layered process that starts with locks on individual rooms and extends to controlled building and floor access, front-desk procedures, cameras, and exterior security patrols.

He also cited tailgating as a problem the university is attempting to address through new technology.

With the Alcatraz system, an enrolled resident approaching an entrance can be authenticated as someone permitted to enter. Plus, it can also detect when another person follows the authorized user through the entrance. This can provide security personnel with information about a potential tailgating event.

Alcatraz says it can detect tailgating in real time and share this information with the existing physical access control infrastructure.

Biometrics supplement campus cards and mobile credentials

For the Kentucky pilot, facial authentication is an add-on credential rather than a replacement for the university's existing identification technologies.

Students can continue to use either their physical Wildcard or mobile ID for building access. These non-biometric methods also remain a requirement for front-desk authorization and access to individual student rooms.

If the facial authentication system fails to recognize an enrolled student, the resident can simply use an existing credential to enter the building. With modern biometric systems, however, such failures have become rare.

Privacy and student choice emphasized

Though biometrics have become commonplace in everyday life, questions surrounding privacy and data management still exist and must be addressed.

Based on the UK Police Department’s description of the project, they are attempting to this in a transparent manner.

They emphasize that participation in the pilot is voluntary. They also stress that biometric information collected through the program is used solely to verify access to the building and is not used to track student movements or behavior.

In an authentication scenario, the system is attempting to confirm that an enrolled individual is the person associated with an authorized credential or biometric profile. It is a one-to-one match.

To further address privacy concerns, both Police Chief Monroe and Alcatraz deliberately describe the technology as facial authentication.

In an authentication scenario, the system is attempting to confirm that an enrolled individual is the person associated with an authorized credential or biometric profile. It is a one-to-one match.

This is a significant delineation from facial recognition systems designed to identify unknown individuals. In this case, an image of a person’s face is captured, converted to a biometric template, and compared against massive data sets. This is a one-to-many match and tends to evoke greater privacy concerns.

For Kentucky, the Woodland Glen IV test provides an opportunity to evaluate those issues alongside the more traditional access control questions of security, reliability and convenience. Learnings from the project should help UK – and other institutions – determine if facial authentication should have a larger role on campus.