Application of machine vision to traffic monitoring and control

Author

Inigo, Rafael M.

Author_Institution

Dept. of Electr. Eng., Virginia Univ., Charlottesville, VA, USA

Volume

38

Issue

3

fYear

1989

fDate

8/1/1989 12:00:00 AM

Firstpage

112

Lastpage

122

Abstract

Research efforts have been made or are in progress on the application of machine vision to traffic monitoring and control. Of all the proposed systems, the wide-area detection system (WADS) appears to be the most reliable. However, limited field tests indicate that some malfunctions exist. Among these are false vehicle detection caused by sudden changes in road illumination and also by shadows or glare on the road produced by automobiles in adjacent lanes. After a brief review of WADS characteristics and problems, algorithm improvements to alleviate these problems are proposed. It is shown that the WADS vehicle detection algorithm becomes more reliable if the vehicle detection threshold includes a term depending on actual road intensity. Finally, an algorithm that would permit vehicle tracking for distances up to 120 m using available charge-coupled-device (CCD) cameras is proposed

Keywords

automotive electronics; computer vision; computerised monitoring; CCD cameras; false vehicle detection; machine vision; road intensity; traffic control; traffic monitoring; vehicle tracking; wide-area detection system; Automobiles; Charge coupled devices; Charge-coupled image sensors; Condition monitoring; Lighting; Machine vision; Roads; Testing; Vehicle detection; Vehicles;

fLanguage

English

Journal_Title

Vehicular Technology, IEEE Transactions on

Publisher

ieee

ISSN

0018-9545

Type

jour

DOI

10.1109/25.45464

Filename

45464