Facial recognition software for identification of powered wheelchair users

Tewkesbury, Giles, Lifton, Samuel, Haddad, Malik, Sanders, David and Gegov, Alex (2022) Facial recognition software for identification of powered wheelchair users. In: Intelligent Systems and Applications: Proceedings of the 2021 Intelligent Systems Conference (IntelliSys) Volume 1.

This is the latest version of this item.

Abstract

The research presented in this paper investigates the use of facial recognition software as a potential system to identify powered wheelchair users. Facial recognition offers advantages over other biometric systems where wheelchair users have disabilities. Facial recognition systems scan an image or video feed for a face, and compare the detected face to previously detected data. This paper reviews the software development kits and the libraries available for creating such as systems and discusses the technologies chosen to create a prototype facial recognition system. The new prototype system was trained with 262 identification pictures and confidence ratings were produced from the system for video feeds from twelve users. The results from the trials and variance in confidence ratings are discussed with respect to gender, presence of glasses and make up. The results demonstrated the system to be 95% efficient in its ability to identify users.

Available Versions of this Item

Actions (login required)

Edit Item Edit Item