Face Morph Detection using PRNU

Abstract

Recently, face morphing has been identified as a major threat to the security of passports and other ID documents. "The magic passport" allows two or more users to be positively identified based on a morphed facial image contained in the passport. We have proposed to use statistical properties of the PRNU to discriminate real from morphed facial images and show the successful application even under post-processing attacks.

Biography

Andreas Uhl is a professor of computer science at the University of Salzburg where he heads the multimedia signal processing and security lab and is deputy head of department. His research interests include biometrics (with emphasis on iris, fingerprints, and vascular biometrics), biometric security (template protection, presentation attack detection, morphing detection), media security (visual data encryption, watermarking and robust hashing) and medical imaging (analysis of gastro-intestinal endoscopic video as well as brain MRI data).