Road-crossing assistance by traffic flow analysis

Adi Perry, Nahum Kiryati

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We present a system to alert visually impaired pedestrians of vehicles approaching a road-crossing without traffic control. The system is computationally efficient, requires low-cost hardware, and can be mounted on existing street infrastructure, such as sign or lighting poles. The incoming video stream, showing the approaching traffic, is transformed to a one-dimensional signal, that is forwarded to a decision module. Preliminary experimental results indicate promising probabilityof- detection and false alarm rates, while providing sufficiently early warning to the pedestrian. The planned target hardware is a solar-charged low cost Android device.

Original languageEnglish
Title of host publicationComputer Vision - ECCV 2014 Workshops, Proceedings
EditorsCarsten Rother, Lourdes Agapito, Michael M. Bronstein
Pages361-374
Number of pages14
ISBN (Electronic)9783319161983
DOIs
StatePublished - 2015
Event13th European Conference on Computer Vision, ECCV 2014 - Zurich, Switzerland
Duration: 6 Sep 201412 Sep 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8927

Conference

Conference13th European Conference on Computer Vision, ECCV 2014
Country/TerritorySwitzerland
CityZurich
Period6/09/1412/09/14

Keywords

  • Approaching traffic analysis
  • Blind
  • Optic flow
  • Resource-limited system
  • Road crossing assistance
  • Visually impaired

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Road-crossing assistance by traffic flow analysis'. Together they form a unique fingerprint.

Cite this