Measurement of Road Traffic Parameters Based on Multi-Vehicle Tracking

TL;DR

This paper presents a computer vision-based system for measuring road traffic parameters using multi-vehicle tracking, employing LBP features, GAB classifiers, and EKF for vehicle detection, classification, and counting.

Contribution

The paper introduces a novel traffic measurement system combining LBP features, GAB classifiers, and EKF tracking, improving accuracy and efficiency over background subtraction methods.

Findings

The proposed system achieves higher accuracy in vehicle detection.

It demonstrates faster execution time compared to background subtraction.

The system effectively tracks and counts vehicles in traffic videos.

Abstract

Development of computing power and cheap video cameras enabled today's traffic management systems to include more cameras and computer vision applications for transportation system monitoring and control. Combined with image processing algorithms cameras are used as sensors to measure road traffic parameters like flow volume, origin-destination matrices, classify vehicles, etc. In this paper we propose a system for measurement of road traffic parameters (basic motion model parameters and macro-scopic traffic parameters). The system is based on Local Binary Pattern (LBP) image features classification with a cascade of Gentle Adaboost (GAB) classifiers to determine vehicle existence and its location in an image. Additionally, vehicle tracking and counting in a road traffic video is performed by using Extended Kalman Filter (EKF) and virtual markers. The newly proposed system is compared with a system based on background subtraction. Comparison is performed by the means of execution time and accuracy.