Modeling and Classification of Vehicles Using Magnetometers
Background
The data collected by urban planning and development bodies reveal that a great deal of resources are wasted because of the road traffic congestion. The number of man hours wasted due to traffic delay and the amount of pollution are significantly huge. Therefore, there is a great demand for intelligent traffic systems which are capable of monitoring traffic to reduce delay and to smoothen the flow of vehicles.
Our Research
In this work, we present two novel approaches of modeling magnetic signature of a vehicle captured using magnetometer. In the first approach, we fit a magnetic dipole model to the magnetic signature, whereas in the second approach we fit a sensor-based model to the magnetic signature. Next, we extract the magnetic dipoles and magnetic rotation angle as features, respectively. Finally, we use support vector machine to classify cars of different lengths.
Publications
- G. V. Prateek, K. Nijil, and K. V. S. Hari, “Classification of vehicles using magnetic field angle model,” 4th International Conference on Intelligent Systems, Modelling and Simulation, Bangkok, 2013, pp. 214-219. [pdf] [slides]
- G. V. Prateek, V. Rajkumar, K. Nijil, and K. V. S. Hari, “Classification of vehicles using magnetic dipole model,” TENCON IEEE Region 10 Conference, Cebu, 2012, pp. 1-6. [pdf] [slides]