Spatiotemporal vehicle tracking: the use of unsupervised learning-based segmentation and object tracking

Author

Chen, Shu-Ching ; Shyu, Mei-Ling ; Peeta, Srinivas ; Zhang, Chengcui

Author_Institution

Sch. of Comput. Sci., Florida Int. Univ., Miami, FL, USA

Volume

12

Issue

1

fYear

2005

fDate

3/1/2005 12:00:00 AM

Firstpage

50

Lastpage

58

Abstract

In this paper, a framework for spatiotemporal vehicle tracking using unsupervised learning-based segmentation and object tracking is presented. An adaptive background learning and subtraction method is proposed and applied to two real-traffic video sequences to obtain more accurate spatiotemporal information on the vehicle objects. As demonstrated in the experiments, almost all vehicle objects are successfully identified through this framework.

Keywords

automated highways; computer vision; image segmentation; image sequences; object detection; road traffic; unsupervised learning; background learning method; image segmentation; intelligent transportation system; object tracking; real-traffic video sequence; spatiotemporal vehicle tracking; subtraction method; unsupervised learning; Convergence; Image segmentation; Iterative algorithms; Partitioning algorithms; Robot vision systems; Robotics and automation; Spatiotemporal phenomena; Tellurium; Traffic control; Vehicles;

fLanguage

English

Journal_Title

Robotics & Automation Magazine, IEEE

Publisher

ieee

ISSN

1070-9932

Type

jour

DOI

10.1109/MRA.2005.1411419

Filename

1411419