In this paper the authors describe a computer vision-based traffic monitoring system able to detect individual vehicles in real-time. Their fully integrated system first obtains the main traffic variables: counting, speed and category; and then computes a complete set of statistical variables. The objective is to investigate some of the difficulties impeding existing traffic systems to achieve balanced accuracy in every condition; i.e. day and night transitions, shadows, heavy vehicles, occlusions, slow traffic and congestions. The system they present is autonomous, works for long periods of time without human intervention and adapts automatically to the changing environmental conditions. Several innovations, designed to deal with the above circumstances, are proposed in the paper: an integrated calibration and image rectification step, differentiated methods for day and night, an adaptive segmentation algorithm, a multistage shadow detection method and special considerations for heavy vehicle identification and treatment of slow traffic. A specific methodology has been developed to benchmark the accuracy of the different methods proposed. The paper is organized as follows. They start out in section 2 with a complete analysis of the traffic scenario. Next, they proceed to describe the different computer vision steps in section 3. Calibration and image rectification are presented in section 4. Segmentation follows in section 5. Detection and tracking is described in section 6. Given its relevance, shadow removal deserves a full section in section 7. Results and implementation issues are discussed in section 8. Finally, they end the paper with the conclusions and suggestions for further improvements.
An adaptive, real-time, traffic monitoring system
Ein adaptives Echtzeit-Verkehrsüberwachungssystem
Machine Vision and Applications ; 21 , 4 ; 555-576
2010
22 Seiten, 20 Bilder, 6 Tabellen, 35 Quellen
Aufsatz (Zeitschrift)
Englisch
An adaptive, real-time, traffic monitoring system
British Library Online Contents | 2010
|Adaptive Background Estimation for Real-Time Traffic Monitoring
British Library Conference Proceedings | 2001
|REAL-TIME TRAFFIC MONITORING METHOD USING CONNECTED CAR AND REAL-TIME TRAFFIC MONITORING SYSTEM
Europäisches Patentamt | 2019
|