Fraunhofer develops a system to address face morphing attacks in border control

The Fraunhofer Institute of Computer Graphics Research (IGD) is working to address security concerns related to face morphing attacks, particularly in border control scenarios. The initiative is part of the comprehensive ATHENE research program, which focuses on biometrics and identity management, with a specific emphasis on countering face morphing attacks.

The current biometric systems employed at border controls are designed to accommodate certain variations in an individual’s appearance. While these systems are trained to tolerate such changes to avoid false rejections of legitimate users, they also become susceptible to morphing attacks.

In an interview, Dr. Naser Damer, a senior researcher at Fraunhofer IGD, said, “It has been trained to accept some changes to the face. If it didn’t, there would be problems. For example, your passport is valid for 10 years, and they expect that the border officials and the machine will be able to identify you with this passport and the corresponding picture during this time.”

In the case of a morphed passport photo, the system may identify both individuals represented in the morphed image as legitimate matches, as the morphed photo contains acceptable elements of each face.

This poses a risk, as individuals whose facial features are included in a morphed passport photo could potentially pass through automated border controls undetected, thereby threatening biometric identification systems and the country’s national security.

The research team at Fraunhofer IGD has implemented two strategies to tackle the issue: developing generalized detection algorithms and conducting simulated attacks.

These algorithms are designed to anticipate morphing attacks using previously unknown methods. Researchers utilize generative adversarial networks to create face morphing attacks and understand their mechanisms.

Fraunhofer is also developing algorithms to evaluate the quality of passport photos before they are approved for official documents. According to the company, low-quality images are more vulnerable to alterations.

During the “ISO and the New ICAO Passport 2025 Standards” workshop, Dr. Andreas Wolf discussed the future-proofing of biometric identity documents by incorporating data elements such as multiple images, 3D images, quality information PAD, and capture device information.

“We cannot claim that these solutions are perfect. The reason for this is simple: there will continue to be new types of attacks in the future, and for this reason, our solution cannot be perfect. But we must work to create a near-perfect solution,” Damer adds.

The ATHENE research center is also involved in various other projects, including image quality control for forensic analysis and passport verification, identity management for face morphing and presentation attacks, and biometrics in embedded systems with limited computational resources.

Fraunhofer IGD will present the research at the it-sa Expo&Congress in Nuremberg, scheduled from October 22 to 24. Additionally, Dr. Damer will participate in a panel discussion at the European Association for Biometrics Research Project Conference 2024, focusing on the role of generative AI and synthetic data.

As part of the EU-funded iMARS research project, the team conducted a second workshop to present its findings on face morphing. While instances of the attack appear to be limited, iMARS has surveyed 20 agencies from 19 countries, including passport offices, border crossings, federal police, ministries, and others. The questionnaire is still open for responses.

Article Topics

biometric identification  |  biometric passport  |  biometrics  |  biometrics research  |  EAB  |  EAB 2024  |  face biometrics  |  face morphing  |  morphing attack