Abstract

We present a traffic monitoring tool, based on computer vision techniques, which is able to perform vehicle tracking and speed control. It uses background substraction techniques like Gaussian modelling or Stauffer & Grimson, morphological operators and stabilization algorithms to process the video, and advanced tracking methods which use Kalman filtering and mean-shift algorithms.

Demo

This demo shows how vehicle tracking is performed using a test sequence. Background substraction is performed using Stauffer & Grimson model.

In this demo we use Gaussian modelling for the background estimation.

Vehicle tracking in Barcelona

This video corresponds to a road of Barcelona.
Background segmentation using Stauffer & Grimson.

Evening vehicle tracking

Our system is also adapted to work at low light conditions.
Background segmentation using Stauffer & Grimson.

The team

This project has been developed by four students of the Master in Computer Vision of the Universitat Autònoma de Barcelona.

4th Workshop on Road Traffic Monitoring

Here you can check others teams' work.