Tracking all traffic: Computer vision algorithms for monitoring vehicles individuals, and crowds

Benjamin Maurin, Osama Masoud, Nikolaos P. Papanikolopoulos

Research output: Contribution to journalArticlepeer-review

35 Scopus citations

Abstract

A vision-based system for monitoring crowded urban scenes is proposed. The approach combines an effective detection scheme based on optical flow and background removal that can locate vehicles, individual pedestrians, and crowds. The detection phase is followed by the tracking phase that tracks all the detected entities. Potential applications include intersection control, traffic data collection, and even crowd control after athletic events.

Original languageEnglish (US)
Pages (from-to)29-36
Number of pages8
JournalIEEE Robotics and Automation Magazine
Volume12
Issue number1
DOIs
StatePublished - Mar 2005

Bibliographical note

Funding Information:
This work was supported in part by the Minnesota Department of Transportation and in part by the National Science Founda tion through grants CMS- 0127893 and IIS-0219863.

Keywords

  • Computer vision algorithms
  • Crowds
  • Intersections
  • Monitoring systems
  • Tracking schemes
  • Traffic objects
  • Vision-based system

Fingerprint

Dive into the research topics of 'Tracking all traffic: Computer vision algorithms for monitoring vehicles individuals, and crowds'. Together they form a unique fingerprint.

Cite this