The Fraunhofer Institute is actively developing solutions to counteract face morphing attacks that pose risks to border control systems. This research, part of the ATHENE program, involves creating detection algorithms and assessing passport photo quality to prevent unauthorized passage. Recent discussions highlight the challenges of achieving a flawless detection system due to the evolving nature of morphing techniques. Upcoming presentations at industry conferences will showcase these efforts and findings.
The Fraunhofer Institute for Computer Graphics Research (IGD) is enhancing security protocols to combat face morphing attacks, particularly in the context of border control systems. This initiative is part of the ATHENE research program, which prioritizes advancements in biometrics and identity management, with a dedicated focus on mitigating the risks associated with face morphing by employing innovative detection methodologies. Current biometric systems at border checkpoints are designed to accommodate natural variations in an individual’s facial appearance over time. This adaptability, while essential for preventing false rejections, inadvertently increases vulnerability to face morphing attacks. In a statement, Dr. Naser Damer, a senior researcher with Fraunhofer IGD, remarked, “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.” Morphing attacks exploit this tolerance; a passport photo altered to blend features of two individuals may lead the system to incorrectly recognize both as legitimate, which can facilitate unauthorized passage through automated controls, posing serious risks to national security. To combat these challenges, the team at Fraunhofer IGD has pioneered two main strategies: the creation of universal detection algorithms and the simulation of morphing attacks. These advanced algorithms are engineered to predict and counteract morphing techniques that have not been previously encountered. Utilizing generative adversarial networks (GANs), researchers are able to produce highly sophisticated face morphing attacks, which in turn enhance their understanding of this threat. In addition to attack detection, Fraunhofer is making strides in improving the quality assessment of passport photographs before acceptance for official use. They assert that images lacking in quality are more easily modifiable and thus more susceptible to morphing. During the “ISO and the New ICAO Passport 2025 Standards” workshop, Dr. Andreas Wolf emphasized the importance of enhancing biometric identity documents with additional data elements, including multiple and 3D images, quality assurance information (pad), and metadata regarding the photographic capture device. Dr. Damer further noted, “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.” The ATHENE research center is also engaged in various other complementary endeavors such as forensic image quality control, passport verification protocols, and the integration of biometrics within embedded systems with limited processing capabilities. These findings will be showcased at the upcoming it-sa Expo&Congress in Nuremberg, scheduled for October 22 through 24, and Dr. Damer is set to feature in a panel discussion at the European Association for Biometrics Research Project Conference 2024, where the implications of generative AI and synthetic data will be a focal point. Furthermore, as part of the EU-funded iMARS research project, the team has hosted follow-up workshops to disseminate insights regarding face morphing techniques and has engaged with 20 agencies across 19 countries to gather further data, with responses still being collected from various stakeholders, including passport offices and border security agencies.
Face morphing attacks present a unique challenge to biometric security systems, particularly in border control settings. In these scenarios, biometric systems must decipher a range of natural changes in appearance caused by aging, lifestyle, or environmental factors. Consequently, these systems are explicitly designed to recognize legitimate users even when their appearance varies over a span of years. However, such flexibility also opens the door for malicious actors to employ morphing techniques—merging the facial features of two different individuals into a single photo that could be fraudulently used as identification. Therefore, technology that can accurately detect and counter these specific threats is pivotal in securing national borders and protecting identity management frameworks. The ATHENE program at Fraunhofer IGD represents a robust research effort dedicated to addressing these sophisticated security challenges through advanced detection techniques and predictive algorithms.
Fraunhofer IGD is taking significant steps to combat face morphing attacks, which threaten border control biometrics by allowing individuals with altered identities to gain unauthorized access. By developing advanced algorithms capable of detecting morphing with high reliability and improving the quality assessment of passport photographs, the initiative seeks to enhance the effectiveness of biometric identification systems. Continuous innovations and the integration of newer data elements will be critical as attackers develop new strategies, underscoring the need for ongoing research and adaptation in biometric security practices.
Original Source: www.biometricupdate.com
