Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/129851
Title: Physical Biometric Spoofing of 3D Dynamic On-Air Signature
Authors: Quintana Hernández, José Juan 
Ferrer, Miguel A. 
Diaz, Moises 
Carmona-Duarte, Cristina 
UNESCO Clasification: 33 Ciencias tecnológicas
Keywords: 3D Automatic Gesture Recognition
3D Biometric Authentication
3D Synthesis Of Human Movements
Kinematic Theory Of Rapid Movements
Sigma-Lognormal Model
Issue Date: 2023
Conference: 56th International Carnahan Conference on Security Technology, ICCST 2023 
Abstract: 3D on-air signatures represent a novel behavioral biometric that is highly challenging to imitate, even from a distance. However, the emergence of robotic arms capable of mimicking human movements poses a potential security vulnerability. This paper examines the possibility of generating 3D fingertip movements with a robotic arm that accurately mimic a 3D on-air signature to circumvent a biometric system. The robotic arm controls trajectory and velocity to generate a human-like movement. Robotic-made signatures are then used to spoof a 3D on-air dynamic signature biometric system, which relies on a DTW verifier. Experiments demonstrate that robotic signatures can effectively bypass a verifier exclusively trained with real signatures, underscoring the importance of incorporating an additional security layer, such as a liveness detector, into these biometric schemes.
URI: http://hdl.handle.net/10553/129851
ISBN: 9798350315875
ISSN: 1071-6572
DOI: 10.1109/ICCST59048.2023.10474253
Source: Proceedings - International Carnahan Conference on Security Technology[ISSN 1071-6572], (Enero 2023)
Appears in Collections:Actas de congresos
Show full item record

Page view(s)

92
checked on Oct 5, 2024

Google ScholarTM

Check

Altmetric


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.