Face-Based Graphical Authentication System Using Hybrid Images


Khaled, Assim Sulaiman and Kam, Yvonne Hwei Syn (2021) Face-Based Graphical Authentication System Using Hybrid Images. In: International Conference on Advances in Cyber Security (ACeS), 8-9 December 2020, Penang, Malaysia.

[img] Text
Face-Based Graphical Authentication System Using Hybrid Images.pdf
Restricted to Repository staff only

Download (2MB)


Authentication systems used in the security world vary greatly but one system, the textual password authentication system, has dominated and become the most common authentication method used today. However, secure textual passwords are difficult to remember. In contrast, faces are easier to remember than text [1]. In this study, a face-based graphical authentication system that utilizes hybrid images has been developed. To the best of the authors’ knowledge, hybrid images of faces has not yet been used for authentication, thus this study uses optical illusion on faces for the purpose of authentication. The main purpose of the method is to reduce the limitations found in conventional text (or PIN) based authentication methods, i.e. passwords are difficult to remember and vulnerability to shoulder surfing attacks. The system was tested on 20 users who logged in while being shoulder surfed. The results showed that shoulder surfing attacks were successfully prevented. With only one attempt, 35% of the users logged in correctly. They had a 72% accuracy in identifying the pass faces. In contrast, none of the shoulder surfers managed to login despite being given three attempts. They guessed correctly only 17% of the user’s pass images.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Image authentication
Subjects: T Technology > TA Engineering (General). Civil engineering (General) > TA1501-1820 Applied optics. Photonics
Divisions: Faculty of Engineering (FOE)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 08 May 2021 14:41
Last Modified: 08 May 2021 14:41
URII: http://shdl.mmu.edu.my/id/eprint/8681


Downloads per month over past year

View ItemEdit (login required)