Real-Time Video Based Highway Traffic Measurement and Performance Monitoring

This paper presents a real-time highway monitoring system for tracking and classification of vehicles with the computation of traffic flow parameters from live video streams. The proposed system robustly detects and tracks vehicles during daylight hours and accurately classifies them into 8 different types by leveraging tracking information. The system is able to process video continuously over long time periods, accumulating large volumes of tracking data to build daily highway models consisting of the traffic flow parameters, density, flow, and speed. These daily models are used to categorize the speed profile of live traffic.

I. Introduction

Highway traffic management requires up-to-date data delivered in real-time along with historical data of traffic conditions to design effective control strategies. In California, inductive loop sensors deliver lane counts and occupancy every 30 seconds from locations all over the state. Berke-ley's Freeway Performance Measurement Project (PeMS) [1] collects and analyzes data from over 26,000 individual lane detectors from 7 of the 12 California Department of Transportation (Caltrans) districts, investigating assorted performance measures. While this large infrastructure is already in place, it is not ideal. Only 60% of CA loop detectors supply usable data. Since the loop detectors must be installed in the road, adding and maintaining them is expensive and intrusive. The failure rate coupled with the cost of maintenance make alternative data collection methods desirable.