One shot learning approach to identify drivers

Malik Haddad, David Sanders, Martin Clifford Langner, Giles Tewkesbury

Research output: Chapter in Book/Report/Conference proceedingConference contribution

49 Downloads (Pure)


This paper presents a new approach to identify users of shared powered mobility platfroms using One Shot learning algorithm. An electronic circuit is created using a Raspberry Pi and a camera module. Python programming language is used to create a program to control the function of the camera module and conduct One Shot learning to identify users. A user interface containing Start and Stop buttons is created to operate the camera module. If Start is pressed, the Python program will trigger the camera to take a snapshot, then the program will identify the face in that snapshot and perform facial recognition to compare that face with a database of user facial images. If a match was found a message box popped-up showing the user name. Once a user is identified, their settings and preferences can be downloaded to the shared powered mobility platform. If the face in the snapshot did not match any of the user facial images in the data base, a message popped-up showing that user was not found in the database. Practical testing showed the system behaved satisfactorily and successfully detected users. One Shot learning allowed new users to be added to the database without the need to retrain the whole system.
Original languageEnglish
Title of host publicationProceedings of SAI Intelligent Systems Conference
Subtitle of host publicationIntelliSys 2021: Intelligent Systems and Applications
EditorsK. Arai
Number of pages8
ISBN (Electronic)9783030821937
ISBN (Print)9783030821920
Publication statusEarly online - 4 Aug 2021
EventIntelliSys 2021 - Amsterdam, Netherlands
Duration: 2 Sept 20213 Sept 2021

Publication series

NameLecture Notes in Networks and Systems
ISSN (Print)2367-3370


ConferenceIntelliSys 2021


  • camera
  • wheelchair
  • disabled
  • one shot learning
  • Python
  • RCUK
  • EP/S005927/1


Dive into the research topics of 'One shot learning approach to identify drivers'. Together they form a unique fingerprint.

Cite this