On User Involvement in Production of Images Used in Visual Authentication
Authors: Karen Renaud

Date: February 2009
Publication: Journal of Visual Languages & Computing, Volume 20, Issue 1
Page(s): 1 - 15
Publisher: Elsevier Science
Source 1: http://dx.doi.org/10.1016/j.jvlc.2008.04.001 - Subscription or payment required

Abstract or Summary:
Recognition-based visual authentication schemes use a variety of different kinds of images. These mechanisms have now matured sufficiently that we should start considering tailoring and fine-tuning themólooking at ways to make them more efficient. Since these mechanisms use images, the obvious starting point in this tailoring process is to consider the image type or genre being used by the mechanism.

Images have a number of properties which are bound to influence the efficacy of the visual authentication mechanism. In this paper the notion of essential and tuning image properties is proposed. The former are those that an image must exhibit or possess in order to be used in visual authentication at allófailure to meet these metrics should disqualify the image from use. Tuning properties, on the other hand, are properties that will improve the efficiency of the mechanism. The tuning property which is the focus of this paper is the user's involvement in the production of his/her secret images.

A longitudinal study was carried out with a visual authentication system in order to determine the effectivity of images with three levels of user involvement, using randomly issued images from an archive, a set of hand-drawn images called doodles, and user-provided photos.

The hand-drawn doodles performed better than both system-issued images and personal photos. Furthermore, whereas doodles demonstrate viability, personal photos have many insuperable problems which make them unsuitable for use in a security setting.

Do you have additional information to contribute regarding this research paper? If so, please email siteupdates@passwordresearch.com with the details.

<-- Back to Authentication Research Paper Index

[Home] [About Us] [News] [Research]

Copyright © 2019 PasswordResearch.com