PROJECT TITLE :

Empirical Study of Face Authentication Systems Under OSNFD Attacks - 2018

ABSTRACT:

Face authentication has been widely available on smartphones, tablets, and laptops. As various personal pictures are published in on-line Social Networks (OSNs), OSN-based facial disclosure (OSNFD) creates significant threat against face authentication. We build the primary try to quantitatively measure OSNFD threat to real-world face authentication systems on smartphones, tablets, and laptops. Our results show that the proportion of vulnerable users that are subject to spoofing attacks is high, that is about sixty four percent for laptop users, and ninety three percent smartphone/tablet users. We tend to investigate liveness detection ways in the important-world face authentication systems against OSNFD threat. We discover that below protection of liveness detection, the proportion of vulnerable pictures is eighteen.eight p.c, however the proportion of vulnerable users is as high as seventy three.three p.c. This proof suggests that the current face authentication systems are not sturdy enough beneath OSNFD attacks. Finally, we develop a risk estimation tool primarily based on logistic regression, and analyze the impacts of key attributes of facial pictures on the OSNFD risk. Our statistical analysis reveals that the foremost influential attributes of facial pictures are image resolution, facial makeup, occluded eyes, and illumination. This tool can be used to evaluate OSNFD risk for OSN pictures to extend users' awareness of OSNFD.