Mobility aids detection using Convolution Neural Network (CNN)

Mukhtar, Amir; Cree, Michael J.; Scott, Jonathan B.; Streeter, Lee V.

doi:10.1109/IVCNZ.2018.8634731

Files

mobility-aids-mukhtar.pdf

Accepted version, 1.129Mb

DOI

10.1109/IVCNZ.2018.8634731

Citation

Export citation

Mukhtar, A., Cree, M. J., Scott, J. B., & Streeter, L. V. (2018). Mobility aids detection using Convolution Neural Network (CNN). Presented at the International Conference on Image and Vision Computing New Zealand (IVCNZ), Auckland, New Zealand: IEEE. https://doi.org/10.1109/IVCNZ.2018.8634731

Permanent Research Commons link: https://hdl.handle.net/10289/12430

Abstract

The automated detection of disabled persons in surveillance videos to gain data for lobbying access for disabled persons is a largely unexplored application. We train You Only Look Once (YOLO) CNN on a custom database and achieve an accuracy of 92% for detecting disabled pedestrians in surveillance videos. A person is declared disabled if they are detected in the close proximity of a mobility aid. The detection outcome was further categorised into five classes of mobility aids and precision was calculated.

Date

2018

Type

Conference Contribution

Publisher

IEEE

Rights

This is an author’s accepted version of an article published in the Proceedings of International Conference on Image and Vision Computing New Zealand (IVCNZ). © 2018 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

Collections

Science and Engineering Papers [2694]