University of Twente Student Theses


Morphing robust face recognition

Szabó, Krisztián (2022) Morphing robust face recognition.

[img] PDF
Abstract:The increased accessibility of face morphing software poses a threat to face recognition systems due to their vulnerability to face morphing attacks. An attacker can merge a picture of themselves with another similar but distinctlooking person, and current face recognition systems would consider the picture to match both persons. After creating an ID with such picture, the ID would no longer be considered unique. This research aims to combat such attacks by creating a shape-free representation for face recognition and measuring its accuracy against other open-source face recognition software. Using the PUT face database, 50 peoples faces have been morphed and warped to generate 4 datasets, to find out if warping could help with improving facial recognition systems by making them robust against morphing attacks.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science BSc (56964)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page