A Fuzzy Set-based Approach for Matching Hand-Drawing Shapes of Touch-based Gestures for Graphical Passwords

TL;DR

This paper introduces a fuzzy set-based method to improve matching accuracy of hand-drawn gestures in graphical passwords, addressing inherent inaccuracy issues in touch-based gesture recognition.

Contribution

It proposes a novel two-dimensional fuzzy set approach utilizing fuzzy cued click points to better match user gestures with stored references, enhancing gesture acceptance.

Findings

Improved gesture matching accuracy demonstrated.

Enhanced user acceptance of gesture-based authentication.

Flexible handling of gesture inaccuracy in touch interactions.

Abstract

This paper presents a two-dimension fuzzy set based approach for matching touch-based gestures using fuzzy cued click point technique. The pro posed approach aims mainly to improve the acceptance of the most closed inac curate hand drawn gestures generated by the user compared with a predefined referenced gesture value that is stored in the user profile. Commonly, gestures are used in order to facilitate the interactive capabilities between humans and computerized systems. Unfortunately, most of current gesturing techniques don't deal at the same level of inaccuracy of gesturing, resulted from the nature of hu man fingers and hands movements. This paper aims, in a more flexible manner, to tackle the inaccuracy problem existed with gesture-based interactions between humans and a computerized system.