Towards Implicit Visual Memory-Based Authentication
Abstract
Selecting and remembering secure passwords puts a high cognitive burden
on the user, which has adverse effects on usability and security.
Authentication schemes based on implicit memory can relieve the user of
the burden of actively remembering a secure password. In this paper, we
propose a new authentication scheme (MooneyAuth) that relies on
implicitly remembering the content of previously seen Mooney images.
These images are thresholded two-tone images derived from images
containing single objects. Our scheme has two phases: In the enrollment
phase, a user is presented with Mooney images, their corresponding
original images, and labels. This creates an implicit link between the
Mooney image and the object in the user's memory that serves as the
authentication secret. In the authentication phase, the user has to
label a set of Mooney images, a task that gets performed with
substantially fewer mistakes if the images have been seen in the
enrollment phase. We applied an information-theoretical approach to
compute the eligibility of the user, based on which images were labeled
correctly. This new dynamic scoring is substantially better than
previously proposed static scoring by considering the surprisal of the
observed events. We built a prototype and performed three experiments
with 230 and 70 participants over the course of 264 and 21 days,
respectively. We show that MooneyAuth outperforms current implicit
memory-based schemes, and demonstrates a promising new approach for
fallback authentication procedures on the Web.
Origin : Files produced by the author(s)
Loading...