![]() ![]() When checked against the photo stored on file at the passport issuance office, the morphed face resembles ID2 closely enough to result in the issuance of a passport. This morphed facial image is submitted alongside ID2's passport application. To highlight how this process might work, imagine the following scenario: A fraudster (ID1) who is not legally able to obtain a passport morphs a photo of their face with the face of another person who is legally able to obtain a passport (ID2). Face morphing involves digitally combining images of two (or more) individuals to create a manipulated image that maintains a likeness to each of the original identities. Most recently, this issue has intensified through the increased availability of advanced image manipulation software that enables fraudsters to create morphed facial images. A common strategy for obtaining a FOG passport is through an accomplice, who holds a genuine passport, submitting a renewal application with the photo of a similar-looking fraudster. These FOG passports are real documents that are issued by an official and can therefore circumvent the previously mentioned anti-counterfeit measures. In light of these anti-counterfeit measures, fraudsters have started to find ways to obtain fraudulently obtained but genuine (FOG) passports ( ITW Security Division, 2017 Middleton, 2014). In response, passport anti-counterfeit measures were developed, making such attempts easy to detect ( UK HM Passport Office, 2020) (e.g., patterns that are visible only under specific artificial illumination). Traditionally, it was common for fraudsters to attempt to create fraudulent identifications by producing fake passports or removing and replacing the photo in a real passport. Here, we extend this previous literature by examining human and computer-based detection of face morphing using a diverse set of facial images to generate high-quality morphs.Īttempts to obtain fraudulent identity documents are not new. The face databases used in these previous studies, however, have a number of limitations, most notably low-quality morphs and/or limited diversity in terms of the race, gender, and age of the faces used to create the stimuli. Early research exploring human detection of morphed images indicates that people frequently accept both low-quality and high-quality morphed images as genuine ( Robertson et al., 2017 Robertson, 2018 Kramer et al., 2019). One relatively new type of fraud that border control agencies are facing is the use of morphed passport photos. The difficulty of this task leaves identity verification processes vulnerable to fraudulent attacks. Much research, however, has shown that matching pairs of unfamiliar faces is a difficult task ( Bruce et al., 1999 Megreya & Burton, 2007 Burton et al., 2010), including for trained identification-checkers ( White et al., 2014). ![]() We frequently rely on photo-based identity documents to verify identity in critical settings such as border control. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |