Traffic monitoring based on real-time image tracking

Ching Po Lin*, Jen Chao Tai, Kai-Tai Song

*Corresponding author for this work

Research output: Contribution to journalConference articlepeer-review

38 Scopus citations


This paper presents a study on a stand-alone image tracking system for automatic traffic monitoring. The proposed image tracker consists of three parts: an edge detection module, an image tracking module and a traffic monitoring module. The edge-detection module is a special designed circuit board, which features fast CCD image processing and feature extraction. Frame rate (60Hz) edge detection of an image with resolution of 320 × 240 pixels is obtained on-board. The image tracking module performs the vehicle tracking in real time. Adopting active contour models and Kalman filtering techniques, we successfully achieved real-time image tracking of multi-lane moving vehicles. The traffic-monitoring module determines the traffic information from the tracked locations of vehicles. The current system provides three types of traffic information: the velocity of multi-lane vehicles, the number of vehicles and car accident detection. Experimental results are presented to demonstrate the real-time tracking of vehicles on an urban artery.

Original languageEnglish
Pages (from-to)2091-2096
Number of pages6
JournalProceedings - IEEE International Conference on Robotics and Automation
StatePublished - 9 Dec 2003
Event2003 IEEE International Conference on Robotics and Automation - Taipei, Taiwan
Duration: 14 Sep 200319 Sep 2003

Fingerprint Dive into the research topics of 'Traffic monitoring based on real-time image tracking'. Together they form a unique fingerprint.

Cite this