Vehicular Mechanical Condition Determination and On Road Traffic Density Estimation Using Audio Signals

Abstract

In this paper we are going to estimate the vehicular traffic density by using acoustic or sound signals. Here we will estimate three probable conditions of traffic that is heavy flow traffic (0-10km/h), medium flow (20-40km/h), and free flow (above 40km/h) traffic. Cumulative sound signals consist of various noises coming from various part of vehicles which includes rotational parts, vibrations in the engine, friction between the tires and the road, exhausted parts of vehicles, gears, etc. Noise signals are tire noise, engine noise, engine-idling noise, occasional honks, and air turbulence noise of multiple vehicles. These noise signals contains spectral content which are different from each other, therefore we can determine the different traffic density states and mechanical condition of vehicle. For example, under a free-flowing traffic condition, the vehicles typically move with medium to high speeds and thereby produces mainly tire noise and air turbulence noise. Here we will use SVM and ANN classifiers. In ANN, we are going to use Feed Forword Network.